Introduction
By June of 2019, Asia had approximately 2.3 billion internet users, making up half (50.7%) of all internet users worldwide [
1]. Within Asia, China was the country with the highest percentage of internet users (37.1%) among its population [
2]. By March of 2020, there were 904 million internet users in China, and the largest percentage of users (21.5%) were those between 20 and 29 years of age, or the young adult population [
3]. The current internet coverage rate was 64.5% in China [
3], still lower than the rates in countries such as the United States and Japan [
2]. It is expected that, with increased internet coverage, the number of internet users in China would likely rise. As internet use becomes more prevalent, the likelihood of excessive use has also increased.
Griffiths and other scholars have recognized the addictive potential of technology use as early as the 1990s, and in a series of papers, they discussed whether excessive internet use can be deemed as an “addiction” [
4,
5]. Yet, the literature on excessive internet use has evolved considerably since, more recent view about excessive internet use seems to be that, rather than a stand-alone addiction, excessive internet use, as problematic internet use (PIU), is more of an umbrella term depicting a range of behaviors conducted via the internet, such as video gaming, shopping, social networking, pornography viewing, cyber-bullying, etc. [
6]. Thus, it might be helpful to differentiate “generalized internet addiction” from “specific internet addiction” [
7].
Consistent with this view, Gaming Disorder (GD), either online or offline, is now included in the 11th revision of the International Classification of Diseases (ICD-11) developed by the World Health Organization (WHO) [
8]. It depicts a pattern of gaming behavior (i.e., digital or video gaming) featuring impaired control over gaming, increased priorities given to gaming over other interests or activities, continuation or escalation of gaming in spite of negative consequences [
8]. To be diagnosed, the behavior has to be severe enough to cause significant impairments in personal, social, occupational, or other functional domains lasting for 12 months [
9]. Earlier, the American Psychological Association (APA) proposed that Internet Gaming Disorder (IGD), a sub-type of PIU, should be included under Conditions Warranting More Experience and Research in the 5th edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-V) [
10]. Some have argued that the conceptualization of IGD under the DSM-V might lack clarity as IGD was considered to include both internet gaming as well as non-Internet computerized gaming despite having the word “Internet” in its nomenclature [
7].
Despite the debate on conceptual differences, two meta-analytic studies reported the pooled prevalence of PIU in the college student population in China to be approximately 11% [
11,
12], higher than the average rates reported in western countries [
13] Meanwhile, playing video or computer games is a common and popular leisure activity among college students worldwide including in China. The latest statistics indicated that, online game companies had the largest share of the market among all listed companies in China, while 532 million or over half (58.9%) of internet users in China were also internet game users [
3]. For some students, gaming is as much part of their everyday life as studying and eating. Compared with the amount of research on PIU, there are fewer studies examining problematic gaming among young adults in mainland China. Several studies examining problematic gaming among university students in Macao and Hong Kong reported varied prevalence, ranging from 14.8% among respondents in Macao to 25.7% among respondents in Hong Kong [
14,
15]. The prevalence of problematic gaming also varied from 2.1% among college students in Beijing to 7.41% among college students in Luzhou, Sichuan [
16,
17]. The varied prevalence rate could be attributed to differences in methodologies or instruments utilized in measuring problematic gaming, or a lack of well-designed studies testing measurement invariance of the same instrument [
18].
Both PIU and problematic gaming constitute a growing public health concern among the college student population in China. Compared to high school students, college students are relieved of the tremendous pressure from preparing for and sitting through the Chinese college entrance exam or Gaokao, they typically devote more time to pursuing other non-academic activities. A study comparing health behaviors of adolescents and college students in China found college students reported more screen-time on average than both middle school and high school students [
19]. Another study comparing prevalence of problematic gaming among adolescents and college students found a higher prevalence among college students [
16]. Spending excessive time playing games can be detrimental. Both PIU and problematic gaming have been linked to symptoms of psychopathology. A meta-analytic study found PIU linked to depression, anxiety, attention deficit/hyperactivity disorder (ADHD), obsessive-compulsive symptoms (OCS), and/or hostility/aggression [
20]. Several studies including meta-analytic reviews found problematic gaming associated with similar psychopathologies such as depression, obsessive-compulsive symptoms, ADHD, and anxiety [
21,
22].
Sleep not only plays a critical role in maintaining our physical and mental health [
23], but it is also increasingly recognized for its contribution to various psychopathologies [
20]. More specially, problems with sleep can act as an outcome as well as a cause of mental disorders [
24]. Sleep problems often occur as a result of PIU, it could enhance addictions to internet usage, and mediate further psychiatric morbidities [
25]. Sleep quality typically reflects the overall experience of sleep, while sleep duration or quantity reflects the amount of time an individual spends sleeping. Some evidence seems to suggest that sleep duration such as night time sleep might be a poor indicator for screening PIU, as individuals might compensate for reduced night time sleep duration by taking naps during the day [
26]. In the seven studies (Table
1) that examined mediation models with different orderings of PIU, problematic gaming, sleep problems, and psychological symptoms, majority of the studies assessed the mediating role of sleep quality in the PIU-psychological health link or the problematic gaming-psychological health link [
27‐
29,
32,
33]. For example, Bhandari et al. found sleep quality seemed to be a stronger mediator, mediating 30.9% of the indirect effect of PIU on depressive symptoms, whereas PIU was a weaker mediator, mediating only 16.5% of the indirect effect of sleep quality on depressive symptoms [
28]. Only two studies examined sleep problems as a mediator of the problematic gaming-psychological health relation [
27,
33], and the effect size varied considerably. A review study found insufficient evidence supporting a consistent sleep pathway between problematic gaming and depression [
34]. The lack of supporting evidence has been attributed to the recognition of addictive gaming as a disorder only recently, and a lack of studies using validated instruments to assess problematic gaming as well as sleep quality [
34]. These findings suggested that, the mediating role of sleep quality in the PIU/problematic gaming and psychological health connection warrants further study.
Table 1
Studies assessing the mediating role of sleep, problematic internet use (PIU), problematic gaming, and psychological symptoms
| United States | Adolescents | Social messaging Web surfingTV/movie watching Gaming | Problems falling alseep;Problems staying asleep;Weekday sleep duration | Depressive symptoms |
| Nepal | Undergraduatestudents | Sleep quality | Internet addiction | Depressive symptoms |
Internet addiction | Sleep quality | Depressive symptoms |
| China | Adolescents | PIU | Sleep quality | Psychological symptoms |
| China | Adolescents | PIU | Depressive symptoms | Sleep disturbance |
Depressive symptoms | PIU | Sleep disturbance |
| China | Adolescents | PIU | Sleep disturbance | Suicidal behavior |
Suicidal behavior | Sleep disturbance | PIU |
| Switzerland | Adolescents | Electronic media use before bed | Sleep duration Sleep difficulties | Depressive symptoms |
| United States | Children | Screen time (i.e. playing video games) | Sleep duration | Emotional/behavioralproblems |
Therefore, this study aimed to address this issue by examining the mediating role of sleep quality in the PIU-psychological distress link and the problematic gaming-psychological distress link using widely-used instruments. College students in China were the population under study mainly for two reasons: first, they belong to the age group that reports greater internet usage and problems associated with it [
35]; second, they tend to experience varied levels of sleep problems, with the prevalence rate ranging between 14.3 and 42.6% [
36,
37]. By examining the phenomenon in one of China’s internet-savvy populations, this study may contribute to further understanding of the complex relationship between PIU, problematic gaming, sleep quality, and overall psychological distress, and to aid in the identification of pathways for intervention.
Discussion
This study was the first to examine the mediating role of sleep quality in the PIU- and problematic gaming-psychological distress link among college students in China. Findings revealed that both PIU and problematic gaming was associated with decreased sleep quality and increased psychological distress. The main finding was that, sleep quality explained 23.5% of the total effect of PIU on psychological distress, and only 17.9% of the total effect of problematic gaming on psychological distress. Thus, sleep quality was deemed a meaningful mediator only in the PIU-psychological distress link. This finding implied that, PIU may increase psychological distress directly or indirectly through decreasing sleep quality; whereas problematic gaming had a strong direct negative influence on psychological distress, and the influence can barely be adequately explained via decreased sleep quality.
This finding may be better understood through the following mechanisms. First, existing evidence seems PIU-sleep relation has been well-documented, whereas the problematic gaming-sleep relation seemed to be less consistent. Excessive internet use has been associated with a wide range of sleep problems, such as reduced sleep duration, sleep disorders, insomnia, frequent sleep disturbances, increased fatigue and drowsiness during the day [
54]. A systematic review found that, compared with those without PIU, those with PIU had 2.2 times increased odds (pooled) of having sleep problems, and a 15-min reduction in sleep duration [
54]. In contrast, one systematic review examined both the PIU-sleep and problematic gaming-sleep relations, finding strong evidence for the PIU-sleep relation, but insufficient evidence supporting the problematic gaming-sleep relation [
34]. In another systematic review, 18 out of 21 studies found a significant association between video gaming and sleep outcomes, and only 1 out of 3 studies specifically focusing on sleep quality found a significant association between problematic gaming and sleep quality [
55]. Second, factors other than sleep quality might contribute significantly to the problematic gaming-psychological distress relation. For example, problematic gaming may strongly reflect individual characteristics such as high neuroticism and external locus of control, which were also highly associated with psychological distress but not via pathways related to sleep [
56].
Griffiths has argued that a distinction should be made between addictions on the internet and addictions to the internet, as internet can be used as a medium or vehicle to fuel other obsessive behaviors such as gambling, gaming, and pornography [
7]. More specifically, a distinction should be made about “generalized internet addiction” and “specific internet addiction” [
7]. Following this logic, problematic gaming behavior maybe best conceptualized as a behavior distinct from PIU. This proposition has been empirically tested in a 2014 study using a nationally representative sample of adolescent gamers in Hungary, the authors found PIU was positively associated with online gaming, online chatting, and social networking, while problematic gaming was only associated with being male and online gaming [
57], lending support to the notion that PIU and problematic gaming were distinct entities. In a second study, the authors examined generalized and specific internet use from college students in Germany, Sweden, China, and Taiwan, and found that except for problematic social network use, the correlation between generalized PIU and all other specific types of PIU were substantially lower or non-significant [
58], implying the uniqueness of specific types of PIU. The main finding of the present study that PIU and problematic gaming were differentially related to psychological distress via sleep quality may have lent further support for the conceptual distinction between PIU and problematic gaming.
Findings on sociodemographic covariates in this study may also have implications for PIU and problematic gaming. Consistent with existing research [
59], the mean score of problematic gaming was found to be significantly higher among male students, indicating that more males engaged in problematic gaming than females. This gender difference might be explained by craving-related neural mechanisms underlying problematic gaming, such that the same gaming stimulation may elicit more craving-related activation in males rather than females [
59]. Internet use and problematic gaming have generally been associated with academic under-achievement, yet, some studies suggested that they had little impact on non-pathological users [
60]. In this study, the mean score of PIU and problematic gaming were both found to be highest among those with poor academic achievement (bottom 20% of the class), yet, due to the cross-sectional nature of the study, causal inferences cannot be drawn. Two indicators of respondents’ socioeconomic status were examined in this study: annual family income (direct indicator) and parental education level (proxy indicator). Annual family income was found to be associated with PIU but not with problematic gaming, and the mean score of PIU was highest among students reporting the lowest family income. While both parents’ highest level of education was associated with PIU as well as problematic gaming, and the mean scores of both PIU and problematic gaming were lowest among respondents with college-educated parents. This indicated that parental education level might be a better indicator of PIU and problematic gaming. However, parental education level has been rarely included in studies on PIU and problematic gaming, and findings from only a few studies across different populations offered inconclusive evidence [
61,
62]. In this study, the mean scores of both PIU and problematic gaming were also found to be higher among those that used any type of substance, and highest among those who used at least 2 types of substances in the past year. Problematic internet use, problematic gaming behaviors, and substance use disorder are considered to share many core features and comorbidities, there is also evidence supporting substantial similarities in the neurobiological mechanisms underlying each type of disorder [
63]. Findings of this study confirmed their strong associations among college students in China, suggesting that strategies to prevent or reduce PIU or problematic gaming may include those targeting substance use prevention. However, it remains to be determined whether prevention or treatment efforts intended for both conditions are more effective.
Limitations
Nonetheless, this study had several limitations. First, mediation analysis was performed using cross-sectional data, thus, causal inferences cannot be drawn. However, it is worth to note that the respective etiology of psychological distress, PIU, problematic gaming, and poor sleep quality is likely multifactorial, with complex interplay between a multitude of intrinsic and external factors. Examining these factors or understanding how they would interact overtime was beyond the scope of the current study. Our mediation model echoed the developing consensus in the literature that problematic gaming may perhaps be a distinct concept from PIU, as reflected in the growing international recognition of IGD rather than PIU as a mental disorder. Nonetheless, results of our study warrant the need for further longitudinal studies to substantiate the temporal sequence of these associations. Second, in the present study, problematic gaming was assessed by the GAS-7, which was developed precedes the DSM-V, and may suffer from incomplete coverage of diagnostic criteria of IGD outlined in ICD-11 or DSM-V [
64]. However, the GAS-7 is considered a screening tool, and was found to be the most common tool used in prevalence as well as longitudinal studies to assess problematic gaming, and exhibited strong psychometric properties [
64]. Further studies utilizing instruments consistent with ICD-11 classification are needed nonetheless. Third, sleep quality was measured using self-report PSQI, which might be subject to recall bias or social desirability bias. Future studies may benefit from comparing findings using both subjective and objective measures of sleep quality.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.