Moderating Effects of Depressive Symptoms and Self-Control Trait on the Association Between Problematic Internet Gaming and Time Spent

Abstract

This study was performed to examine whether depressive symptoms (DPs) and low self-control (LSC) traits has modifying effects on the association of time spent on Internet gaming and problematic Internet game use (PIGU). Among 1,535 middle school students, 42.1 percent reported playing ≥1.5 hours of Internet games/day on average, and 10.0 percent were classified as PIGU. Stratified multiple logistic regression analysis showed gameplay of ≥1.5 hours/day led to more PIGU in the non-LSC traits group (higher DP group, p < 0.001; lower group, p = 0.019). In LSC trait group, in contrast, time spent on Internet gamine was not associated with PIGU regardless of the level of DP. Adolescents without LSC traits, to which most youths belong, might have higher possibility of PIGU if they do not limit their time spent on Internet gaming, regardless of whether they are depressed or not. In case of LSC trait, additional strategies would be needed. ClinicalTrials.gov identifier: NCT02415322.

Introduction

Compared with traditional video games, Internet games allow for a higher level of interaction between users and gaming systems.¹ Additional factors such as betting and irregular rewards and significant improvements in gaming accessibility, as a result of advances in information and communication technology, have cultivated an environment that promotes addictive behaviors. Furthermore, adolescents are more vulnerable to addiction than adults because of their particular stage of brain development.^2,3 As a testament to these facts, the prevalence of adolescents having problems of Internet gaming disorder is estimated to range from 1.2 percent to 10.7 percent.^4,5

Excessive Internet gaming could alter brain function,³ having potential adverse effects on psychosocial health, such as addiction, cognitive development and learning, social development, and aggressive behavior.⁶ Moreover, disproportionate Internet gaming may heighten physical health problems, such as obesity, eye and visual symptoms, and musculoskeletal problems.⁷ Apart from the adverse impact Internet gaming has on mental health in adulthood, the impact on adolescents' health outcome itself is sufficient to accentuate problematic Internet game use (PIGU) as an important public health issue.^8–10

Addictive disorders could be established only when the condition of apparent exposure to risk factor exists. For example, Internet gaming disorder (IGD) is only diagnosed after an individual has spent a certain amount of time playing Internet games; however, it has been reported that the amount of time spent playing Internet games would not be sufficient to predict IGD.^11,12 In other words, some individuals would experience IGD after a same amount of gameplay, whereas others would not.

To describe such differences, it is needed to consider the variables reported in previous studies as factors that increase the risk of Internet gaming disorder. The most frequently considered variables are depression and low self-control (LSC): given the theoretical explanations of addictive behavior,^13–15 depression is one of the characteristics associated with emotional vulnerability, and self-regulation is regarded as one of the characteristics associated with impulsive trait. Yu et al.¹⁶ found that online gamers with lower self-control were more likely to be problematic game users, and King et al.¹⁷ found depression was associated with problematic gaming. Other researchers^11,18 reported relationship between online game playtime and the problematic use. However, how all three variables relate to problematic gaming was not made clear.

Therefore, we examined relationship between the amount of time adolescents spent playing Internet games and problematic Internet gaming under the condition of depressive symptoms (DPs) and self-control traits status. We hypothesized that DPs and LSC status would moderate the association between how much time an individual spent playing Internet games and PIGU.

Methods

Data collection

Among 2,319 students enrolled in the Internet user Cohort for Unbiased Recognition of Gaming Disorder in Early Adolescents (iCURE) study being undertaken in South Korea,¹⁹ the baseline assessment data collected from 2015 to 2016 was used in this study.

Participants were selected from schools, and informed consent and assent were obtained from the parents and students who agreed to participate, respectively. In the iCURE study, baseline assessment was performed by asking students to complete a self-reported, web-based questionnaire. Their parents underwent a one-on-one interview or completed a self-reported questionnaire depending on the content of the questionnaire. Academic performance and mother's final educational attainment variables analyzed in this study was measured by parents' self-report.

Individuals met criteria for inclusion in this study if they were (a) a middle school student and (b) had played an Internet game within the past year. Individuals were excluded if information on the time they spent playing Internet games was lacking. A total of 1,920 students of the 2,319 enrolled in the iCURE cohort were middle school students, and 1,587 of them had used an Internet game within the past year. After excluding individuals for whom the abovementioned data could not be obtained, 1,535 students were included in the final analysis. As a whole, 99.9 percent were seventh graders (first year in middle school; 11- to 12 years old).

Measures

Internet gaming use-elicited symptom screen

Problematic Internet gaming was assessed by the Internet Gaming Use-Elicited Symptom Screen (IGUESS), developed based on the diagnostic criteria of IGD suggested in the DSM-5. The IGUESS is a nine-item self-reported scale and found to be reliable and diagnostically valid.²⁰ The Cronbach's alpha was 0.846 in this study. The IGUESS includes nine questions, all of which reflect IGD diagnostic criteria. Respondents were asked to rate their answers to each question using a four-point Likert scale; 0 (not at all) to 3 (always). A minimum score of 10 or higher (score range: 0–27) was defined as problematic Internet gaming.

Time spent on Internet gaming

To calculate the time each participant spent playing Internet games, students were asked to indicate the average amount of time they engaged in gameplay per day. Students were then asked to specify the number of weekdays (of 5 days) and weekends (of 2 days) they played Internet games. Total weekly playtime was calculated and categorized into two groups: greater or equal to or less than 1.5 hours per day on average. According to the result of a preliminary analysis by the authors, the diagnostic accuracy of PIGU was highest when the cutoff point of average time spent on Internet gaming was 1.5 hours per day than 2 hours per day. Therefore, in this study, Internet game use was divided into two groups: 1.5 hours or more or less.

Depressive symptoms

DPs were evaluated using the Children's Depression Inventory (CDI),²¹ which is a self-administered depression screening tool used worldwide. The CDI comprises 27 items, and asks respondents to choose from one of the three categories that best describes themselves within the past 2 weeks. Each item is rated from 0 to 2, with a total score range of 0–54. A higher score indicates a higher level of DP. The Cronbach's alpha of the CDI was 0.898 in this study. To analyze interaction effects of the DP, those in the upper 25 percent of the total score were defined as the group with high (depressive symptoms positive, DP+) and low (depressive symptoms negative, DP−) levels of DP, respectively.

Self-control status

LSC traits were investigated using the Self-Control Rating Scale (SCRS).²² This instrument measures the degree to which an individual pursues immediate satisfaction via self-report. Each item is rated from 1 to 4, and a high score indicates impulsive tendencies and having strong inclinations to pursue immediate satisfaction. The Cronbach's alpha of this instrument was 0.820 in this study. To analyze the interaction effects of LSC traits, total SCRS scores were divided into upper 25 percent (LSC trait positive, LSC+) and the others (LSC trait negative, LSC−).

In addition, confounding variables such as gender, subjective academic performance (excellent, good, average, or needs improvement), and maternal educational attainment (high school graduate or lower, college, university or higher) were included in the analysis according to the result of literature review. In addition, physical or musical leisure activities was defined as participating in sports (tennis, badminton, squash, soccer, football, or basketball) or playing musical instrument at least twice a week and at least 30 minutes at a time.

Regarding the game genre used recently, up to five game titles that had been enjoyed online for the past 3 months were listed by self-report, and the titles of those games were classified into game genres such as role-playing game (RPG; World of Warcraft, Maple Story, Dungeon & Fighter, etc.), shooting game (Sudden Attack, Counter Strike, Guns Online, etc.), simulation game (League of Legends, StarCraft, etc.), and others. The cases where one or more game(s) of each genre was engaged were defined as the users of each game genre.

All the variables were collected through the self-report of the participants, except for the mother's final education attainment and the participant's academic performance at which the data were collected through the self-report of the participants' parent or guardian.

Analysis

Data were analyzed using the SAS 9.3 statistical software program (SAS Institute, Inc., Cary). Descriptive analyses were conducted using frequency and proportion. χ² tests were conducted to analyze unadjusted relationship between each variable of interest and outcome variables. When the p value was <0.1 in χ² test, the variable was included in the univariate and multivariate logistic regression model.

The association between the average time spent engaging in Internet gameplay per day and PIGU was analyzed for each of the four groups combined with DP and self-control traits via univariate and multivariate logistic regression analyses. An adjusted odds ratio (OR) and a confidence interval (95% CI) were estimated for each multiple logistic regression model.

Results

Sample characteristics and association with PIGU

Participant characteristics of 1,535 and the distribution of PIGU in relation to these characteristics are given in Table 1. About 68 percent of the participants were male, and 42.1 percent of participants spent 1.5 hours or more per day playing Internet games on average. In the four groups, which combined DP and self-control levels, 60.0 percent of the subjects were negative for both DP and LSC. In turn, the subject who were positive for both accounted for 12.1 percent.

Table 1.

General Characteristics of 1,535 Subjects

Variable	Category	n (%)
Gender	Female	493 (32.1)
Gender	Male	1042 (67.9)
Academic performance	Good or excellent	684 (44.6)
Academic performance	Average or needs improvement	851 (55.4)
Mother's final educational attainment^a	High school graduation or lower	532 (34.7)
	College	356 (23.2)
	University or higher	634 (41.3)
	No response	13 (0.8)
Physical and/or musical leisure activities	No	334 (21.8)
Physical and/or musical leisure activities	Yes	1,201 (78.2)
RPG game use	No	1,143 (74.5)
RPG game use	Yes	392 (25.5)
Simulation game use	No	858 (55.9)
Simulation game use	Yes	677 (44.1)
Shooting game use	No	1,070 (69.7)
Shooting game use	Yes	465 (30.3)
Average time spent on Internet game use	<1.5 h/day	889 (57.9)
Average time spent on Internet game use	≥1.5 h/day	646 (42.1)
LSC and DP^a	LSC− and DP−	921 (60.0)
	LSC− and DP+	199 (13.0)
	LSC+ and DP−	230 (15.0)
	LSC+ and DP+	185 (12.1)
Problematic Internet game use^b	No	1382 (90.0)
Problematic Internet game use^b	Yes	153 (10.0)
Total		1,535 (100.0)

LSC+, low self-control positive, which means upper 25 percent of the Self-Control Rating Scale reversed score, ≥46 (range 20–71); LSC–, low self-control negative; DP+, depressive symptom positive, which means upper 25 percent of the Children's Depression Inventory score, ≥12 (range 0–54); DP–, depressive symptom negative.

Internet Game Use-Elicited Symptom Screen (GUESS) score ≥10.

RPG, role-playing game.

Stratified χ² test showed the frequency of PIGU was significantly higher in the subjects who spent 1.5 hours or more per day on Internet gaming in all groups except the group in which the subjects were LSC+ and DP− (LSC− DP− group, p = 0.002; LSC− DP+, p < 0.001; LSC+ DP−, p = 0.112; and LSC+ DP+, p = 0.048). Among the confounding variables, the following variables showed p value of <0.1: (a) LSC− DP− group: academic performance (p = 0.001), and physical and/or musical leisure activities (p = 0.053); (b) LSC− DP+ group: physical and/or musical leisure activities (p = 0.003), and RPG game use (p = 0.051); (c) LSC+ DP− group: none; and (d) LSC+ DP+ group: sex (p = 0.077), physical and/or musical leisure activities (p = 0.007), and simulation game use (p < 0.05) (Table 2).

Table 2.

Frequency of Problematic Internet Gamers in Four Groups by the Status of Depressive Symptom and Self-Control: Results of χ² Test (N = 1,535)

Variable/category	N (%)
	Frequency of problematic Internet game user ^a in each subgroup^b
	LSC− DP− (n = 921)	p	LSC− DP+ (n = 199)	p	LSC+ DP− (n = 230)	p	LSC+ DP+ (n = 185)	p
Gender
Female	5 (1.8%)	0.155	11 (11.3%)	0.280	10 (20.4%)	0.173	19 (26.8%)	0.077
Male	23 (3.6%)		17 (16.7%)		23 (12.7%)		45 (39.5%)
Academic performance
Good or excellent	6 (1.6%)	0.001	12 (14.5%)	0.894	10 (13.5%)	0.804	15 (33.3%)	0.838
Average or needs improvement	22 (5.0%)		16 (13.8%)		23 (14.7%)		49 (35.0%)
Mother's final educational attainment^c
High school graduation or lower	11 (3.9%)	0.564	13 (19.7%)	0.281	12 (11.8%)	0.219	33 (39.8%)	0.230
College	5 (2.3%)		4 (9.5%)		11 (22.0%)		12 (25.0%)
University or higher	12 (2.9%)		11 (12.6%)		10 (13.0%)		18 (34.6%)
Physical and/or musical leisure activities
No	9 (5.4%)	0.053	12 (27.9%)	0.003	9 (15.0%)	0.867	30 (47.6%)	0.007
Yes	19 (2.5%)		16 (10.3%)		24 (14.1%)		34 (27.9%)
RPG game use
No	18 (2.6%)	0.128	15 (10.9%)	0.051	23 (13.3%)	0.427	42 (32.6%)	0.377
Yes	10 (4.6%)		13 (21.3%)		10 (17.5%)		22 (39.3%)
Simulation game use
No	18 (3.4%)	0.515	16 (15.0%)	0.699	18 (15.3%)	0.687	25 (26.0%)	0.011
Yes	10 (2.6%)		12 (13.0%)		15 (13.4%)		39 (43.8%)
Shooting game use
No	20 (2.9%)	0.769	16 (11.8%)	0.169	17 (12.9%)	0.461	38 (31.1%)	0.170
Yes	8 (3.3%)		12 (19.0%)		16 (16.3%)		26 (41.3%)
Average time spent on Internet game use (hour per week, mean ± SD)	9.97 ± 9.86		11.52 ± 12.74		15.01 ± 12.74		17.48 ± 17.09
<1.5 h/day	10 (1.7%)	0.002	6 (5.0%)	<0.001	11 (10.4%)	0.112	21 (26.6%)	0.048
≥1.5 h/day	18 (5.4%)		22 (27.5%)		22 (17.7%)		43 (40.6%)
Total	28 (3.0%)		28 (14.1%)		33 (14.3%)		64 (34.6%)

IGUESS score ≥10.

LSC+, low self-control positive, which means upper 25 percent of the Self-Control Rating Scale reversed score, ≥46 (range, 20–71); LSC–, low self-control negative; DP+, depressive symptom positive, which means upper 25 percent of the Children's Depression Inventory score, ≥12 (range, 0–54); DP–, depressive symptom negative.

Missing cases were excluded in the analysis (n = 13).

IGUESS, Internet Gaming Use-Elicited Symptom Screen.

Stratified multiple logistic regression analyses by DP and LSC trait

As a result of stratified multiple logistic regression analyses for each of the four groups combined by DP and LSC traits level, it was found that those who spent 1.5 hours or more playing Internet games per day had a significantly higher OR for PIGU (OR = 2.60; 95% CI, 1.17–5.70; p = 0.019) compared with those who spent <1.5 hours on Internet gameplay in the group with lower level of self-control traits and DP. Furthermore, in the group with lower self-control traits but a higher level of DP, a significantly greater OR for PIGU (OR = 6.45; 95% CI, 2.45–17.21, p < 0.001) was shown among the adolescent who spent 1.5 hours or more playing Internet games per day than those who spent <1.5 hours on Internet gameplay (Table 3).

Table 3.

Results of Stratified Multiple Logistic Regression Analysis

Subgroup^a	Variable	Adjusted OR^*	p
LSC– DP–	Academic performance
	Good or excellent	1.00
	Average or needs improvement	3.48 (1.38–8.11)	0.008
	Physical and/or musical leisure activities
	No	1.00
	Yes	0.53 (0.23–1.22)	0.135
	Average time spent on Internet game use
	<1.5 h/day	1.00
	≥1.5 h/day	2.60 (1.17–5.70)	0.019
LSC– DP+	Physical and/or musical leisure activities
	No	1.00
	Yes	0.33 (0.13–0.82)	0.016
	RPG game use
	No	1.00
	Yes	1.48 (0.60–3.61)	0.385
	Average time spent on Internet game use
	<1.5 h/day	1.00
	≥1.5 h/day	6.45 (2.45–17.21)	<0.001
LSC+ DP–	Average time spent on Internet game use
	<1.5 h/day	1.00
	≥1.5 h/day	1.86 (0.86–4.05)	0.116
LSC+ DP+	Gender
	Female	1.00
	Male	1.55 (0.76–3.14)	0.226
	Physical and/or musical leisure activities
	No	1.00
	Yes	0.38 (0.19–0.74)	0.004
	Simulation game use
	No	1.00
	Yes	2.05 (1.05–4.01)	0.036
	Average time spent on Internet game use
	<1.5 h/day	1.00
	≥1.5 h/day	1.66 (0.86–3.21)	0.133

OR, odds ratio.

On the contrary, the group with LSC traits did not show a significant association between ≥1.5 hours of time spent playing Internet games and PIGU, irrespective of the level of DP.

Figure 1 provides the result of stratified multiple logistic regression analysis to visually check the association of time spent on Internet game use and PIGU in each group. As shown, in the case of lower level of LSC, expressed as LSC negative, ORs of PICU were significantly increased with ≥1.5 hours time spent on Internet gaming regardless of the level of DP, and especially, the effect was stronger in the DP+ group than in the negative group.

FIG. 1.

Odds ratios of problematic Internet game use with ≥1.5 hours Internet gaming compared with <1.5 hours per day in four subgroups combined by self-control and depressive symptom level. Based on the result of χ² test, adjusted for (A) academic performance status, and doing physical exercise or musical activities; (B) RPG game use, and doing physical exercise or musical activities; (C) no variable needed adjusted; and (D) gender, simulation game use, and doing physical exercise or musical activities; LSC+, low self-control positive, which means upper 25 percent of the Self-Control Rating Scale reversed score, ≥46 (range, 20–71); LSC–, low self-control negative; DP+, depressive symptom positive, which means upper 25 percent of the Children's Depression Inventory score, ≥12 (range 0–54); DP–, depressive symptom negative. RPG, role-playing game.

However, in the case of higher level of LSC, expressed as LSC+, ORs of PICU were not increased with ≥1.5 hours time spent on Internet gaming regardless of the level of DP. In addition, there was little difference in the ORs according to whether the level of DP was positive or negative.

Discussion

This study tested whether DP and LSC traits made different prevalence of PIGU. We also evaluated whether the association between time spent playing Internet games and PIGU varied according to the status of LSC traits and DPs. Our results indicated that there was a positive association between time spent playing Internet games and PIGU, whereby adolescents who played Internet games for 1.5 hours or more per day were more likely to experience PIGU. Such effect was only observed in the lower 75 percent of the LSC trait group. The ORs for PIGU was higher in the students, which showed elevated DPs than in the group with low levels of DPs. For the groups having upper 25 percent of LSC trait, the time spent playing Internet games did not have a significant effect on PIGU, regardless of the degree of DP.

How would the results of this study be explained? It has been frequently reported that an increase in online game use could increase the possibility of IGD.²³ When DPs are accompanied, the reward obtained from gaming would be greater than those without DPs,²⁴ and the effect of increasing game playtime might be further amplified. However, when self-control is lower, a different situation occurs. According to Mill and Allen,²⁵ lower self-control would increase maladaptive motivation for gaming, which might contribute to problematic game use regardless of time spent on gaming. In other words, if self-control is lower, game use itself might be likely to be a matter rather than the amount of gaming.

Then, why is there no difference according to the amount of time spent on gaming, even if DP is added to the lower self-control? It could be explained by previous studies that problematic gaming was predicted by self-control rather than depression when these variables were considered in a multiple analytical model.²⁶

Previous studies have examined the relationship between time spent playing Internet games, DP, or impulsive traits with PIGU. These studies reported that adolescents with high DP,^27,28 high impulsive traits,^29,30 and those who spend prolonged periods of time engaging in Internet gameplay³¹ are more likely to show Internet game-related problems. These studies examined the three above-mentioned variables as adjusted variables for each other.

However, if we consider the fact that DP and LSC traits are coexisting features, and that current knowledge cannot completely agree the possibility that both variables are an independent cause of outcome, DP and LSC traits need to be examined for modification effects than confounding effects to prevent making inappropriate conclusions.³² In this context, our findings, which were obtained by investigating the association between the time spent playing Internet games and PIGU, after setting DP and LSC traits as effect modifiers, are unique.

An important question to address is what makes some adolescents more vulnerable to PIGU when they spend enough time engaging in Internet gameplay. Susceptibility for the other two types of risk factors—emotional vulnerability and biological-impulsive traits, respectively—coexists in one individual, of course, there would be a difference in degree; hence, it would be supposed that the two factors might interact. In other words, adolescents with high DP only (emotional vulnerability) could be self-medicating in the form of excessive Internet game use without other activities,²⁴ as immersing themselves into Internet gaming could hyperstimulate the reward pathway.

Meanwhile, adolescents with LSC trait or high immediate satisfaction-pursuing characteristics (biological vulnerability-impulsivity) seem to have innate traits that exceed their emotional vulnerability and/or the impact of the degree of game exposure. From another perspective, such adolescents may not obsess over the reward derived from Internet games because they also engage in other activities that give them immediate satisfaction.

Regarding the sex difference, which is frequently reported as a factor related to problematic game use, no difference was observed in this study. This is similar to Rehbein et al.'s report,²³ where male sex was not related to video game dependence when other risk factors such as game genres were considered together. In addition, the study population differed from other studies in terms of analyzing subgroups according to self-control and DP. Future studies could use longitudinal design to further confirm sex difference in the association of self-control, DP, and time spent on online gaming with the problematic use.

The findings of this study have important implications for the prevention of PIGU. With excessive Internet gaming emerging as an important health issue for adolescents, many parents have banned Internet gaming from their children. However, there may not be a one-size-fits-all intervention strategy for adolescents. In other words, whereas restricting the amount of time spent on Internet games may be effective for adolescents without LSC traits, the amount of Internet gaming may be less important for adolescents with LSC traits. As given in this study, our findings suggest that interventions to prevent PIGU in adolescents should be tailored for each group.

In addition, our study had several strengths. First, we included all middle school student enrolled in the iCURE study with enough sample size. Second, the variables included in this study were measured using valid and reliable instruments. Furthermore, in an attempt to minimize information bias, data collected from adolescents and their parent were merged and analyzed. This strategy allowed us to collect more reliable data according to the questions: for instance, data on time spent playing Internet games were collected from adolescents, and data on academic performance from parents.

Limitation of this study was that the self-reporting method may have led to us to under- or overestimate our results, which is a common limitation associated with studies utilizing self-reported data.^27,30 However, although not mentioned in this study, the authors validated the association between self-reported data of the time spent playing Internet games by comparison with the actual time spent playing Internet games measured using a smartphone application and found no significant differences statistically between the two methods.

Nevertheless, owing to the cross-sectional nature of this study, we cannot conclusively describe the temporal relationship between Internet game exposure and PIGU outcome. Thus, a prospective cohort study should be performed to shed light on the effects of Internet gameplay duration on PIGU in each of the four combined groups of varying levels of DP and LSC traits.

Footnotes

Authors' Contributions

S.-J.J., H.W.Y., H.J., and H.-K.L. designed the study and wrote the protocol. S.-J.J. conducted the statistical analysis and interpreting the data. S.-J.J. wrote the first draft of the article and all authors contributed to revising the draft and have approved the final version of the article.

Informed Consent

All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (the Institutional Review Board of the Catholic University of Korea (IRB approval number: MC17EESI0080) and with the Declaration of Helsinki of 1975, as revised in 2000. Informed consent and assent were obtained from the parents and children who agreed to participate, respectively, in this study.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This study was supported by a grant from the Korean Mental Health R&D Project, Ministry of Health and Welfare, Republic of Korea (HL19C0048).

References

Fauth-Bühler

, Mann

. Neurobiological correlates of internet gaming disorder: similarities to pathological gambling. Addict Behav, 2017; 64:349–356.

Somerville

, Jones

, Casey

. A time of change: behavioral and neural correlates of adolescent sensitivity to appetitive and aversive environmental cues. Brain Cogn, 2010; 72:124–133.

Potenza

. Biological contributions to addictions in adolescents and adults: prevention, treatment and policy implications. J Adolesc Health, 2013: 52(2 Suppl 2):S22–32.

Petry

, Rehbein

, Ko

, et al. Internet Gaming Disorder in the DSM-5. Curr Psychiatry Rep, 2015; 17:72.

, Chen

, Han

, et al. Prevalence and factors of addictive Internet use among adolescents in Wuhan, China: interactions of parental relationship with age and hyperactivity-impulsivity. PLoS One, 2013; 8:e61782.

Palaus

, Marron

, Viejo-Sobera

, Redolar-Ripoll

. Neural basis of video gaming: a systematic review. Front Hum Neurosci, 2017; 11:248.

Department of Health. (2014) E-report: report of advisory group on health effects of use of Internet and electronic screen products. Hong Kong: Department of Health, Government of the Hong Kong Special Administrative Region. Available at https://www.studenthealth.gov.hk/english/internet/report/files/e_report_wa.pdf (accessed Sep. 23, 2019).

Yau

, Crowley

, Mayes

, et al. Are Internet use and video-game-playing addictive behaviors? Biological, clinical and public health implications for youths and adults. Minerva Psychiatr, 2012; 53:153–170.

American Psychiatric Association. (2013) Diagnostic and Statistical Manual of Mental Disorders. 5th ed. Arlington, VA: American Psychiatric Publishing.

10.

World Health Organization. (2019) International Classification of Diseases-11 for mortality and morbidity statistics. World Health Organization. Available at https://icd.who.int/browse11/l-m (accessed Nov. 11, 2019).

11.

Gentile

. Pathological video-game use among youth ages 8 to 18: a national study. Psychol Sci, 2009; 20:594–602.

12.

Starcevic

, Aboujaoude

. Internet addiction: reappraisal of an increasingly inadequate concept. CNS Spectr, 2017; 22:7–13.

13.

Blaszczynsk

, Nower

. A pathways model of problem and pathological gambling. Addiction, 2002; 97:487–499.

14.

Billieux

, Maurage

, Lopez-Fernandez

, et al. Can disordered mobile phone use be considered a behavioral addiction? An update on current evidence and a comprehensive model for future research. Curr Addict Rep, 2015; 2:156–162.

15.

Lee

S-Y

, Lee

, Choo

. Typology of Internet gaming disorder and its clinical implications. Psychiatry Clin Neurosci, 2017; 71:479–491.

16.

, Mo

, Zhang

, et al. Impulsivity, self-control, interpersonal influences, and maladaptive cognitions as factors of internet gaming disorder among adolescents in China: Cross-sectional Mediation Study. J Med Internet Res, 2021; 23:e26810.

17.

King

, Delfabbro

, Griffiths

. Trajectories of problem video gaming among adult regular gamers: an 18-month longitudinal study. Cyberpsychol Behav Soc Netw, 2013; 16:72–76.

18.

King

, Delfabbro

, Zwaans

, et al. Clinical features and axis I comorbidity of Australian adolescent pathological Internet and video game users. Aust N Z J Psychiatry, 2013; 47:1058–1067.

19.

Jeong

, Yim

, Jo

, et al. Study protocol of the internet user Cohort for Unbiased Recognition of gaming disorder in Early adolescence (iCURE), Korea, 2015–2019. BMJ Open, 2017; 7:e018350.

20.

, Yim

, Lee

, et al. The Internet Game Use-Elicited Symptom Screen proved to be a valid tool for adolescents aged 10–19 years. Acta Paediatr, 2018; 107:511–516.

21.

Kovacs

. The Children's Depression, Inventory (CDI). Psychopharmacol Bull, 1985; 21:995–998.

22.

Robin

, Fischel

, Brown

. The measurement of self-control in children: validation of the Self-Control Rating Scale. J Pediatr Psychol, 1984; 9:165–175.

23.

Rehbein

, Kleimann

, Mössle

. Prevalence and risk factors of video game dependency in adolescence: results of a German nationwide survey. Cyberpsychol Behav Soc Netw, 2010; 13:269–277.

24.

Hussain

, Griffiths

. The attitudes, feelings, and experiences of online gamers: a qualitative analysis. Cyberpsychol Behav, 2019; 12:747–753.

25.

Mills

, Allen

. Self-determination theory, internet gaming disorder, and the mediating role of self-control. Comput Hum Behav, 2020; 105:106209.

26.

Rho

, Lee

, et al. Risk factors for internet gaming disorder: psychological factors and internet gaming characteristics. Int J Environ Res Public Health 2017 Dec 27; 15:40.

27.

Wei

, Chen

, Huang

, et al. The association between online gaming, social phobia, and depression: an internet survey. The association between online gaming, social phobia, and depression: an internet survey. BMC Psychiatry, 2012; 12:92.

28.

Király

, Sleczka

, Pontes

, et al. Validation of the ten-item Internet Gaming Disorder Test (IGDT-10) and evaluation of the nine DSM-5 Internet Gaming Disorder criteria. Addict Behav, 2017; 64:253–260.

29.

Nuyens

, Deleuze

, Maurage

, et al. Impulsivity in multiplayer online battle arena gamers: preliminary results on experimental and self-report measures. J Behav Addict, 2016; 5:351–356.

30.

Choo

, Gentile

, Sim

, et al. Pathological video-gaming among Singaporean youth. Ann Acad Med Singap, 2010; 39:822–829.

31.

Gentile

, Choo

, Liau

, et al. Pathological video game use among youths: a two-year longitudinal study. Pediatrics, 2011; 127:e319–329.

32.

Shrier

, Pang

. Confounding, effect modification and the odds ratio: common misinterpretations. J Clin Epidemiol, 2015; 68:470–474.