This study is part of a large ongoing study on adolescent media usage [65-67] for which ethical approval (no. LPEK-0307) has been obtained by the local psychological ethics commission at the Center for Psychosocial Medicine of the University Medical Center Hamburg-Eppendorf. Verbal informed consent was obtained from both parents and adolescents prior to their participation in the study. Participants were informed of their right to withdraw from the study at any time without any consequences. Compensation was offered for participation in the form of charity vouchers. The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the Declaration of Helsinki of 1975, as revised in 2008.

Data were collected by Forsa, a market and opinion research institute, using computer-assisted telephone interviews (forsa.omniTel). Participants were children and adolescents aged 12-17 years, selected through a multistage systematic random sampling process based on the German Arbeitskreis Markt- und Sozialforschungsinstitute e.V. (ADM; a business association representing private sector market and social research agencies) telephone master sample. Sampling occurred in 2 stages: first, geographically distributed sampling points were identified across Germany; second, households within these points were selected via a random-route method, with 1 eligible adolescent randomly chosen per household. A total of 2075 households were contacted, and after obtaining verbal consent from parents and participants, the final participation rate was 48.2%, yielding a sample of 1000 participants. The dual-frame sampling method (landline and mobile numbers) ensured broad coverage of the population. The sample was stratified by age, gender, and region to align with national demographic distributions based on microcensus data from the Federal Statistical Office of Germany. Data were also collected on educational and occupational status (school student, apprentice, and voluntary service), school attainment (desired or achieved school-leaving certificate), and recent attendance patterns (absences from school, apprenticeship, or work within the preceding 4 weeks). A rolling sample approach was used, meaning that there was no fixed gross sample; instead, previously processed numbers were continuously replaced with new numbers following a systematic call plan. Representativity using this approach has been established in prior research [68-71]. Participants were informed that their participation was voluntary, the data were collected anonymously, and would not be passed on to third parties. Due to the nature of telephone interviews, there were minimal missing responses, which were recorded as NA or “don’t know” responses. A comprehensive description of the data collection methods and demographic characteristics can be found elsewhere [67].

To evaluate IGD symptoms as defined by DSM-V, we used the well-established Internet Gaming Disorder Scale (IGDS). This scale is a single-factor, polythetic tool comprising 9 questions with a binary response format (no/yes), with higher scores reflecting greater risk for IGD. It has been frequently used with German adolescent samples, showing good psychometric properties [67]. For the purposes of the descriptive results, the cutoff indicating disordered gamers is a score of at least 5 [72]. In this sample, the Cronbach α score was 0.58. Potential reasons for this low score are provided in the “Strengths and Limitations” section in the “Discussion” section.

Sleep quality was assessed by administering the 9-item, validated Sleep Reduction Screening Questionnaire (SRSQ) [73]. The SRSQ measures chronic sleep reduction including its associated consequences, affecting everyday life across 9 questions with a 3-step ordinal scale. The SRSQ has been validated and used in various populations, including German adolescents, with good psychometric properties [65,73]. Higher scores indicated more pronounced indications of poor sleep quality, with scores above 17.3 indicating chronic sleep reduction according to the Youden criterion reported elsewhere [65]. In this sample, the Cronbach α score was 0.71.

The Strengths and Difficulties Questionnaire (SDQ) is a widely used screening tool for assessing behavioral and emotional difficulties in children and adolescents as well as a sensitive screener for ADHD-combined subtype [74,75]. In this study, we used the hyperactivity-inattention subscale of the SDQ to assess 2 core components of executive functioning: attentional control (inattention) and behavioral inhibition (hyperactivity or impulsivity). These domains are commonly impaired in individuals with IGD and are also core symptoms of ADHD, which is itself widely recognized as a disorder of executive dysfunction [32,33]. Executive dysfunction, in the context of this study, is operationally defined as difficulty regulating attention and behavior, as measured by self-reported inattentiveness and hyperactivity. While this subscale does not capture the full spectrum of executive functioning, it targets 2 key subdomains most relevant to an IGD population. Responses were rated on a 3-point Likert scale: “not applicable,” “partially applicable,” and “clearly applicable.” Higher scores indicate greater executive functioning problems. This subscale has been used in child psychiatric diagnostic settings and demonstrates satisfactory to good psychometric properties [76-78]. In this sample, the Cronbach α score was 0.62.

We followed recommendations for hypothesizing and constructing mediation models and interpreting effects [79-82]. To examine whether sleep quality mediates the effect of IGD symptoms on executive dysfunction, we used structural equation modeling and a nonparametric bootstrap method with bias-corrected CIs. Figure 1 shows the path diagram of the mediation models. As shown in Figure 1A, we tested the following effects: (1) effect of IGD symptoms on sleep quality (path a), (2) effect of sleep quality on executive function (path b), (3) indirect effect of IGD symptoms on executive functioning through sleep quality (path a×b), (4) direct effect of IGD symptoms on executive functioning, controlling for sleep quality (path c′), and (5) total effect, representing the overall effect of IGD symptoms on executive functioning (path a×b+c). An additional mediation model was run, reversing the predictor and outcome variables (Figure 1B).

Bootstrapping was used because it does not require the indirect effect (a×b) to be normally distributed, making it the preferred method to determine whether the indirect effect is different from zero [82,83]. A significant indirect effect suggests at least partial mediation, meaning that sleep quality explains part of the relationship between IGD symptoms and executive functioning. If the direct effect is nonsignificant in the presence of a significant indirect effect, this indicates that the effect of the predictor on the outcome variable occurs entirely through mediating variables (ie, sleep quality). If the direct effect remains significant, sleep quality only partially explains this relationship (partial mediation).

Software R (version 4.2.2; R Foundation for Statistical Computing) was used for data cleaning and analysis. To account for missing data, multiple imputation was used for each model. We used 60 imputations as recommended by Zhang and Wang [83] for the 21.6% missing data of 1 variable (IGDS) considered to be missing at random. The multiple imputation and mediation analyses were conducted using the bmem package (version 2.1; Comprehensive R Archive Network), which imputes missing data and runs the structural equation modeling in each of the 1000 bootstrap samples that were used in the analysis. To speed up the imputation, we recruited parallel computing with the number of cores set to 8. After obtaining the mediation effect estimates, bias-corrected CIs for the model parameters and mediation effects were constructed. To investigate the nature of the indirect effect, a simple slope analysis was conducted using the ggeffects package (version 1.6.0; Comprehensive R Archive Network), estimating marginal means (EMM) for different levels of sleep quality.

Research has demonstrated a relationship between IGD and executive dysfunction, indicating that IGD is associated with specific cognitive deficits in attention and hyperactivity or impulsivity. Additionally, both have been linked with poor sleep [58,87-89]. However, the role of sleep as a potential mediator in the relationship between IGD and executive dysfunction remains unclear. We proposed that the current treatment models’ failure to consider sleep as a mediating factor limits their efficacy and overlooks a crucial link in this relationship. Therefore, the aim of this study was to investigate how sleep quality is related to IGD and executive dysfunction. Given the high prevalence of IGD among adolescents and the critical role of executive functioning in their development [15,41,43,85], addressing potential negative impacts remains a key priority.

First, in comparing the strength of the parameter estimates of the 2 models (Figure 3), IGD symptoms appeared to be more strongly linked to sleep quality than executive dysfunction. When controlling for sleep quality, the association between executive dysfunction and IGD was stronger than the reverse. In both models, sleep quality partially mediated the relationship between executive dysfunction and IGD. Partial mediation suggests that while sleep quality accounts for some of the variance in executive dysfunction among individuals with IGD, additional factors—such as reward system dysfunction [90], psychological inflexibility [91], and attentional biases [92]—may also contribute to this relationship. The results of this study indicated that sleep quality may be an important mediatory mechanism when considering the relationship between IGD and executive dysfunction. This is especially relevant, given that more than half of the participants who met criteria for IGD had comorbid chronic sleep reduction, consistent with other research findings [53].

Overall, these findings suggest that the relationship between IGD, executive dysfunction, and sleep quality may operate via a reciprocal feedback loop, with each affecting the other. In other words, IGD can produce poor sleep quality, which in turn can lead to worse executive dysfunction, resulting in a feedback loop. This aligns with previous research demonstrating the reciprocal nature of executive dysfunction and IGD [48,93,94], and, novelly, highlights the role of sleep as a significant mediator.

Categorizing the level of sleep quality allowed us to qualify the nature of its mediation. Specifically, we found that at higher levels of sleep quality, IGD symptoms did not significantly predict executive dysfunction. Conversely, at lower levels of sleep quality, the prediction stood. This suggests that good sleep quality appears to mitigate the impact of IGD on executive dysfunction. Therefore, improving sleep quality may potentially buffer against the negative effects of IGD and executive dysfunction. By extension, identifying poor sleep quality early can help target those at greater risk of compounded symptoms.

These findings have several practical and theoretical implications. Previous studies have linked poor sleep status, including reduction in sleep quality, with IGD [53]; however, few current treatment programs include sleep as a factor for improvement [95-98]. Similarly, sleep issues may not always be generally considered in screening procedures. This is despite the evidenced sleep issues among individuals with IGD, including disruptions to the circadian clock [99], increased sleep latency, and decreased total rapid eye movement sleep [100].

From a preventative standpoint, patients presenting with IGD may benefit from a routine assessment of sleep quality during triage, such as the SRSQ or the Sleep Screening Questionnaire Children and Adolescents [101]. Regarding treatment, interventions aimed at improving sleep quality, such as cognitive-behavioral therapy for insomnia, sleep hygiene education, and relaxation techniques, may be particularly beneficial for individuals with IGD and executive dysfunction [102-105]. For instance, programs such as Res@t (Resource-Strengthening Training for Adolescents with Problematic Digital-Media Use) include structured guidance on optimizing sleep routines, such as minimizing screen exposure before bedtime, establishing consistent sleep-wake schedules, and implementing behavioral techniques to reinforce healthier sleep habits [95]. Ensuring adequate and restful sleep may also reduce the risk of developing or exacerbating IGD and executive dysfunction. The bidirectional nature of these relationships underscores the importance of considering multiple factors in management and treatment of IGD.

Ensuring that patients and their families receive psychoeducation regarding the importance of good sleep hygiene and its impact on mental health may help prevent the exacerbation of IGD symptoms and executive dysfunction. Encouraging healthy lifestyle habits that promote better sleep (eg, regular sleep schedules, reduced screen time, physical activity, and blue light filters before bed) can serve as preventive and mitigating measures [106-109]. This can be in addition to psychoeducation regarding the impacts of the sedentary lifestyle, repetitive microtrauma, and obesogenic environment that can accompany excessive gaming and may lead to pathological musculoskeletal injury and a decline in general health status [110].

Finally, impacts on executive functioning are a common finding across addictions. Dysregulated reward processing, diminished impulse control, and aberrant reward-based learning linked to IGD have been similarly found in individuals with gambling disorder [111-114] and substance use disorders [19,44]. Given the partial mediation observed, interventions should consider a multipronged approach that not only addresses sleep hygiene but also directly targets executive function impairments. This ensures that improvements in sleep are supplemented by interventions that target the residual cognitive deficits that persist even when sleep is optimized. Future research could also investigate whether sleep quality, like its role in IGD and executive function, may serve as a potential mediator in other types of addictions.

Whereas previous studies investigating IGD are often hindered by small and heterogeneous samples with a lack of focus on possible mediation [26,48], the large sample size in this study is representative of the general population of children and adolescents. Second, the application of bootstrap structural equation modeling to examine mediation, along with multiple imputation to address missing data, follows the latest recommendations for using advanced statistical methods for mediation models [79-82].

Along with its strengths, there are several limitations to consider in this study. The Cronbach α scores for 2 implemented measures (IGDS and SDQ) were lower than usual. As the alpha coefficient reflects both the properties of the scale and the attributes of the sample, unique characteristics in the sample may have contributed to these lower coefficients [115]. For example, it is possible that because the questionnaire was administered over the phone instead of being self-administered, which is the more typical method, respondents might have answered more cautiously to the interviewers in an effort to appear more socially normative or desirable. Supporting this, Jeong et al [116] found a significant discrepancy between self-reported measurements and clinically verified IGD diagnoses among adolescents, with a false-negative rate of 44%. Thus, while the internal consistency of the scales should be interpreted with some caution, it may also be that effects of interest were understated in this sample, and that future studies may find stronger effects. It should also be noted that these 2 scales are widely and successfully used. For example, the IGDS has good established psychometric properties in similar populations [20,117], and the SDQ, one of the most widely used measures of child mental health globally, has been translated into 80 languages [118]. Although the hyperactivity-inattentive subscale on the SDQ has shown promise as a short and efficient screener for executive dysfunction [75,119], future research should also consider using more comprehensive measures such as full neuropsychological batteries that test a range of executive components. Given that ADHD-related symptoms are typically assessed through teacher or parent reports, future studies should incorporate multi-informant assessments. Additionally, future research should also include objective measures of sleep quality, such as actigraphy or polysomnography, instead of relying solely on self-reports, as was the case in this study. The cross-sectional nature of the data limits the ability to infer longitudinal relationships between variables. Therefore, it remains unclear, for instance, whether executive dysfunction leads to the development of IGD or vice versa; only that a relationship exists between these variables. Future longitudinal studies incorporating multimethod sleep assessments will be needed to establish causality. The generalizability of these results is limited to teenagers. Given the vulnerability of children to IGD and the importance of early intervention [49,63,85], it would be beneficial to determine the strength of the tested mediation models in a younger population.

Recent years have seen a rise in the prevalence of problematic digital media behaviors among adolescents. Executive dysfunction has been associated with IGD, and sleep impairment is also common among adolescents with IGD. The findings of this study align with previous research linking these variables. However, this study’s novelty lies in suggesting that sleep quality can act as a partial mediator between IGD and executive dysfunction. This has significant clinical implications, emphasizing the need to screen for sleep issues as a preventative strategy and to address sleep quality in intervention approaches. Future research is needed to determine the temporal directionality of the relationship between IGD, sleep, and executive functioning incorporating objective sleep measures and comprehensive neuropsychological batteries. Experimental research could also test the effectiveness of sleep-focused interventions in improving cognitive outcomes and reducing IGD symptoms.