This study not only provides regional prevalence data for suicidal ideation, IGD, insomnia, and depression among adolescent Internet gamers in Shanghai, China, but also elucidates the relationships and the underlying mechanisms of these four psychological disorders using a serial multiple mediation model. The prevalence for suicidal ideation in our general study samples and the specific IGD group was 27.2% and 46.9%, respectively. Compared with adolescents with normal Internet use, individuals with IGD were more vulnerable to suicidal ideation. These results accord with those of a previous study of 9,758 students from five European countries(43), which found that 45.86% of students with problematic Internet gaming behaviors showed suicidality. Adolescents with other types of Internet addiction also show a substantial risk of suicidal ideation(18). Female, younger, poorer students in our study showed a greater likelihood of suicidal ideation. These sociodemographic differences echo other findings from different countries(44, 45). One study reported that participants with a median age of 13 years (IQR: 8–15 years)(6) showed harmful consequences of suicidal ideation and suicidal attempts that required clinical treatment.
The present findings identified significant relationships among IGD, insomnia, depression, and suicidal ideation, and strongly suggest that more efforts are needed to monitor Internet use and prevent adolescent insomnia and depression to mitigate the risk of suicidal ideation. Our results show a strong mediating effect (39.0%) of insomnia on the association between IGD and depression. This effect is similar to that found in studies conducted with Hong Kong adolescents(46) and adolescents in Nepal(29), yet much lower than the effect (60.6%) shown for adolescents in Guangzhou, China(27). This large difference may partly reflect the study population. Although both studies targeted secondary school students, the Guangzhou study included all grades of compulsive education, whereas our study examined only the first grade. The former population may have been exposed to greater sleep deprivation owing to the stress of competing for entrance to senior high schools. Barley et al.(47) emphasized that stress is an important influence on the comorbidity of Internet addiction and insomnia. Thus, the relationships among addiction to types of Internet usage other than IGD, and insomnia and depression need to be further explored while considering the influence of different population background characteristics. Besides, our model tested the bidirectional relationship between IGD and insomnia with a statistically insignificant result, and founded that insomnia was a stronger mediator than IGD in the associations among IGD, insomnia and depression, which was in accordance with previous literature.
This is the first study to examine a serial multiple mediation model of the associations between IGD, insomnia, depression, and suicidal ideation in Chinese adolescents. The mediation model demonstrated that IGD was sequentially correlated with insomnia in the first step, and further positively affected the onset of depression, which was associated with a greater risk of suicidal ideation. Additionally, insomnia alone failed to significantly mediate the pathway to suicidal ideation, which highlights the role of depression as a key mediator in the whole model. This finding is in accordance with a study by Sami et al.(48) on the correlation between sleep disturbance and suicidal ideation. It is possible that depression is the strongest risk factor for suicidal ideation in the presence of IGD, insomnia, or other psychological disorders. It is hard to predict suicidal ideation, but the risk factors identified in this study can be measured. Therefore, better treatment for IGD, insomnia, and depression is central to the prevention of suicidal ideation in the Internet era.
This study had some limitations. First, to reduce the questionnaire length and response time, the suicidal ideation measure was a single question with three response options, which is inadequate for accurate diagnoses. A validated theory-based scale is preferable for identifying suicidal ideation in adolescents. Although many other studies(49, 50) have also used a single question (i.e., the ninth item on the PHQ-9 scale, “thoughts that you would be better off dead or of hurting yourself in some way”) to identify suicidality, which was similar to the question used here, a more comprehensive inquiry about suicidal ideation would increase the reliability of outcomes. Second, this was a cross-sectional study, and so the reliability of the findings may be low. Longitudinal studies are needed to confirm a temporal effect of these associations. Third, convenience sampling was used to obtain the study population, so the prevalence of suicidal ideation shown here may not apply to adolescents in other regions, as regional differences may affect prevalence. Finally, in addition to the two mediators tested here, other factors may be important in the pathway from IGD to suicidal ideation. For example, impulsivity is a characteristic trait of adolescence and is strongly associated with addictive disorders (e.g., IGD) and other risky behaviors (e.g., suicidal ideation)(51). As adolescence is characterized by internal psychological change and external interpersonal adaption, future studies should examine other psychological factors that may be associated with suicidal ideation.