Psychiatry Research 228 (2015) 128–135

Contents lists available at ScienceDirect

Psychiatry Research
journal homepage: www.elsevier.com/locate/psychres

Pathological Internet use among adolescents: Comparing gamers

and non-gamers
Esther Strittmatter a,b,1, Michael Kaess a,n,1, Peter Parzer a, Gloria Fischer a, Vladimir Carli c,
Christina W. Hoven d,e, Camilla Wasserman d,f, Marco Sarchiapone f, Tony Durkee c,
Alan Apter g, Julio Bobes h, Romuald Brunner a, Doina Cosman i, Merike Sisask j,
Peeter Värnik j, Danuta Wasserman c
a
Department of Child and Adolescent Psychiatry, University Hospital Heidelberg, Heidelberg, Germany
b
University Medical Center Münster, Department of Child and Adolescent Psychiatry, Psychosomatics and Psychotherapy, Münster, Germany
c
National Centre for Suicide Research and Prevention of Mental Ill-Health (NASP), Karolinska Institutet, Stockholm, Sweden
d
Department of Child and Adolescent Psychiatry, New York State Psychiatric Institute, Columbia University, New York, New York, USA
e
Department of Epidemiology, Mailman School of Public Health, Columbia University, New York, New York, USA
f
Department of Health Sciences, University of Molise, Campobasso, Italy
g
Feinberg Child Study Centre, Schneider Children's Medical Centre, Tel Aviv University, Tel Aviv, Israel
h
Department of Psychiatry, School of Medicine, CIBERSAM, University of Oviedo, Oviedo, Spain
i
Department of Clinical Psychology Department, Iuliu Hatieganu University of Medicine and Pharmacy, Cluj-Napoca, Romania
j
Estonian-Swedish Mental Health and Suicidology Institute (ERSI), Tallinn University, Tallinn, Estonia

art ic l e i nf o a b s t r a c t

Article history: “Internet gaming disorder” was recently included in Section 3 of the Diagnostic and Statistical Manual of
Received 26 November 2014 Mental Disorders (DSM-5). Non-gaming Internet activities were not considered because of a lack of
Received in revised form evidence. This study examined whether gamers differ from non-gamers with respect to their
11 February 2015
psychological well-being among students who show pathological Internet use (PIU). This cross-
Accepted 7 April 2015
Available online 30 April 2015
sectional study was conducted within the project “Working in Europe to Stop Truancy Among Youth
(WE-STAY)”. A total of 8807 European representative students from randomly selected schools were
Keywords: included. The Young Diagnostic Questionnaire was applied to assess PIU, and students with this
Internet addiction condition were divided into gamers (PIU-G) and non-gamers (PIU-NG). Overall, 3.62% and 3.11% of the
Internet gaming disorder
students were classified as having PIU-G and PIU-NG, respectively. A multinomial logistic regression
Psychological impairment
revealed that students with PIU-G and those with PIU-NG showed similarly increased risks for emotional
Adolescents
WE-STAY symptoms, conduct disorder, hyperactivity/inattention, self-injurious behaviors, and suicidal ideation
and behaviors. Students with PIU-G were more likely to be male and have a higher risk for peer
problems than those with PIU-NG. Students with PIU-NG had a higher risk of depression than those with
PIU-G. The significant psychological impairment of PIU-NG suggests that it should be considered in
future diagnostic criteria.
& 2015 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Internet use (PIU). Many authors have investigated PIU in general
(Durkee et al., 2012; Fu et al., 2010), which has also been referred to as
Research has shown the potential harmful effects caused by “Internet addiction” (Young, 1998), “computer addiction” (Wieland,
pathological use of the Internet (Schimmenti et al., 2014a). In the 2005), “compulsive Internet use” (Meerkerk et al., 2006) and “proble-
past, different terms and concepts have been applied to pathological matic Internet use” (Caplan, 2002). Some studies have even specially
focused on Internet gaming (Ko et al., 2005). For this condition, terms
such as “pathological video gaming” (King et al., 2013), “video game
n
Correspondence to: Section for Disorders of Personality Development, Clinic of addiction” (Mößle and Rehbein, 2013), “problematic computer game
Child and Adolescent Psychiatry, Center for Psychosocial Medicine, University of use” (Festl et al., 2013) and “problematic online game use” (Kim and
Heidelberg, Blumenstr.8, 69115 Heidelberg, Germany. Tel.: þ49 6221 566915.
E-mail address: [email protected] (M. Kaess).
Kim, 2010) have been used. Though different terms and definitions
1
Both authors contributed equally; therefore, both should be considered first have been used previously, the construct of PIU implies a pattern of
authors. uncontrolled Internet use resulting in clinical impairment or distress.

http://dx.doi.org/10.1016/j.psychres.2015.04.029
0165-1781/& 2015 Elsevier Ireland Ltd. All rights reserved.
 E. Strittmatter et al. / Psychiatry Research 228 (2015) 128–135 129

In the Young Diagnostic Questionnaire (YDQ; Young, 1998), a widely other studies suggest that applications involving social interactions
used tool to investigate PIU (Bakken et al., 2009; Cao et al., 2007; may also abet the development of PIU (Caplan, 2002; van den
Durkee et al., 2012), five out of eight criteria have to be met for a Eijnden et al., 2008).Petry and O’Brien (2013) stated that additional
diagnosis of PIU. These criteria bear great similarity to the new DSM-5 research is needed to “identify the defining features of the condi-
criteria for “Internet gaming disorder” (see Table 1). This condition of tion” and to “determine prevalence rates in representative epide-
increasing interest was recently included in Section 3 of the Diagnostic miological samples in countries around the world”.
and Statistical Manual of Mental Disorders, 5th Edition (DSM-5) The aim of the study was to empirically contribute to the
released in May 2013 (APA 2013), thus calling for additional research ongoing debate if non-gaming Internet activities should also be
(Petry et al., 2014). According to the American Psychiatric Association included in a future diagnostic category (e.g. in the ICD-11) as this
(APA), Internet gaming disorder is based on 9 criteria: preoccupation has important implications for both prevention and therapy. Thus,
with Internet games, withdrawal symptoms, tolerance, loss of control, the primary focus of the present study was to examine a
loss of interests in previous hobbies and entertainment, continued representative sample of European adolescents to identify indivi-
excessive use despite knowledge of psychosocial problems, dissimula- duals with PIU who engage in online gaming (PIU-G) and those
tion, dysfunctional affect regulation, hazard and loss. Five of these with PIU who engage in online activities other than gaming (PIU-
9 criteria must be present over a 12-month period for diagnosis. The NG). Furthermore, any differences with regard to their comorbid
meetings regarding the next version of the International Classification psychopathologies, engagement in self-harming behaviors and
of Diseases (ICD-11; which is expected to be published in 2015) have psychological well-being were also assessed.
discussed whether a new section called “behavioral addictions” with
the subtopic “computer and Internet addiction” should be included
(Mann et al., 2013). 2. Methods
Furthermore, the condition of an Internet gaming disorder
requires that the symptoms should lead to clinically significant 2.1. Procedures and sample
impairments or distress. Both general PIU and Internet gaming are
significantly associated with comorbid psychopathologies, risk- The present study was conducted within the framework of the
taking and self-harming behaviors, and an overall reduction in “Working in Europe to Stop Truancy Among Youth (WE-STAY)” project
psychological well-being (Caplan, 2002; Gentile et al., 2011; Ha funded by the 7th Framework Program (FP7) of the European Union.
et al., 2007; Kaess et al., 2014; Ko et al., 2009; Lemmens et al., Representative samples were recruited in accordance with previously
2011; Van Rooij et al., 2011). Nevertheless, a lack of studies exists established and validated procedures from the “Saving and Empower-
regarding a specific focus on non-gaming Internet activities or ing Young Lives in Europe” (SEYLE) study (Carli et al., 2013;
comparisons with gaming activities. Wasserman et al., 2010). In each country, a list of all eligible schools
Currently, the new DSM-5 criteria for Internet gaming disorder within the study sites was generated according to specific inclusion
refer exclusively to online (and offline) computer games. The debate and exclusion criteria. Finally, 132 randomly selected schools were
is ongoing as to whether other forms of online activities have a recruited from the following European countries: Estonia, Germany,
similar addictive potential and might be associated with similar Italy, Romania and Spain. Sweden served as the coordination center,
psychological impairments (Schimmenti et al., 2014a). The addictive and experts from Columbia University, New York, United States
potential of different Internet applications was recently assessed in provided methodological consultation. Ethical approval was obtained
a longitudinal study of 447 heavy Internet users (Meerkerk et al., from the local ethics committees at each study site. After a complete
2006). Although the majority of time spent online included e-mail- description of the study was provided to participants and their
ing, downloading, chatting and surfing, the authors found that parents, their written informed consent was obtained. All question-
gaming and searching for online sexual activities were the most naires were administered at the schools and in the official language of
important Internet applications associated with PIU. In contrast, the respective country.

Table 1
Operationalization of the criteria for PIU assessed using the YDQ items and compared with the criteria for Internet gaming disorder in the DSM-5.

PIU (YDQ; (Young, 1998)) Internet gaming disorder (DSM-5)

Defined as: Time spent online for non-academic or non-job (i.e., recreational) purposes Persistent and recurrent use of Internet games leading to clinically significant
impairments or distress
Note: Can involve non-Internet computerized games
Cut-off: Z5
Questions Criteria

Do you feel preoccupied with the Internet (i.e., do you think about previous online Preoccupation with Internet games
activities or anticipate your next online session)?
Do you feel restless, moody, depressed or irritable when you attempt to cut down or stop Withdrawal symptoms when Internet gaming is taken away
Internet use?
Do you feel the need to use the Internet in increasing amounts of time to achieve Tolerance: the need to spend increasing amounts of time engaged in Internet
satisfaction? games
Have you repeatedly made unsuccessful efforts to control, cut back or stop Internet use? Unsuccessful attempts to control participation in Internet games
Do you stay online longer than you originally intended? Loss of interest in previous hobbies and entertainment as a result of (and with
the exception of) Internet games.
Have you jeopardized or risked the loss of significant relationships, jobs, or educational Continued excessive use of Internet games despite knowledge of psychosocial
opportunities because of the Internet? problems
Has jeopardized or lost a significant relationships, jobs, or educational/career
opportunities because of participation in Internet games
Have you lied to your family members, therapist or others to conceal the extent of your Has deceived family members, therapists, or others regarding amount of
involvement with the Internet? Internet gaming
Do you use the Internet as a way of escaping from problems or relieving dysphonic Use of Internet games to escape or relieve a negative mood
moods (e.g., feelings of helplessness, guilt, anxiety or depression)?
130 E. Strittmatter et al. / Psychiatry Research 228 (2015) 128–135

The study sample was composed of 9758 students (55.1% female/ 2.3. Statistical analyses
44.9% male; mean age¼ 15.0 years, S.D.¼ 1.3 years) who completed
the questionnaire; 951 students were excluded because of missing Descriptive analyses were used to determine the rates of
data concerning gender, PIU, or psychosocial variables. Ultimately, normal Internet use, PIU-G and PIU-NG by gender and country,
8807 students were included in the present analyses. The distribu- respectively. Means and standard deviations (SDs) were calculated
tion with regard to the different countries was as follows: 1558 of the for the dimensional variables (BDI-II, DPS-CD, SDQ, and WHO-5).
students came from Estonia, 2564 from Germany, 2117 from Italy, The results are presented as boxplots. For all dimensional variables
1237 from Romania and 1331 from Spain. analysis of variance (ANOVA) was calculated, followed by pairwise
Group comparisons with Tukey's HSD adjustment for multiple
testing. The frequencies of the categorical variables (suicidal and
2.2. Psychological measures self-injurious behaviors) were described for the three Internet user
groups. For each categorical variable, comparisons between the
PIU was assessed using the Young Diagnostic Questionnaire Internet groups were calculated with logistic regressions followed
(YDQ) (Young, 1998). The YDQ has been widely used to investigate by pairwise group comparisons with Sidak's correction for multi-
PIU (Bakken et al., 2009; Cao et al., 2007; Durkee et al., 2012). The ple testing. Following this univariate approach, a multinomial
diagnosis of PIU is based on a pattern of Internet use resulting in regression analysis was calculated using groups of internet use
clinical impairment or distress. Eight criteria are retrospectively (normal, PIU-G, and PIU-NG) as dependent variables; gender, BDI-
evaluated in the YDQ over the last year using “yes” or “no” II, DPS-CD, SDQ-subscales, WHO-5, PSS and DSHI were used as
questions, with a total score ranging from 0 to 8. These criteria explanatory variables. Missing values among the explanatory
bear great similarity to the criteria mentioned in the DSM-5 for variables were replaced with multiple imputed values using
Internet gaming disorders (see Table 1), including both conditions’ multivariate imputation based on the chained equations algorithm
having to meet five criteria for a diagnosis of PIU. In the present to prevent estimation bias due to participant exclusion. The
study, all students were screened with the YDQ for PIU as a first regression was then calculated for each of the 20 imputed
step. Students fulfilling five or more criteria in the YDQ were datasets, and the results were combined. The continuous expla-
classified as having PIU. Students who fulfilled four or less criteria natory variables were standardized to ease the coefficient inter-
in the YDQ were classified as non-PIU. pretation. The regression coefficients are presented in their
Furthermore, the frequency of Internet application use (e.g., for exponential form as relative risk ratios, with respect to normal
gaming, chatting, surfing, and school work) was assessed on a 4- Internet use as base category. All statistical analyses were per-
point Likert scale (“never”, “occasionally”, “frequently” or “always”). formed with Stata 13. A significance level of p¼ 0.05 was chosen.
Only Internet applications used frequently or always were consid-
ered in the statistical analysis. Gaming was defined as playing single
or multi-user online games. Chatting included writing or reading 3. Results
e-mails, visiting chat rooms and using social networks. Surfing
included reading the news, downloading music or videos, watching Of the 8807 students included in the analysis, 4889 (55.5%) were
videos and using other services. In a second step, all students with female, and 3918 (44.5%) were male. The mean sample age was 15.0
PIU were divided into those who frequently engaged in gaming years (S.D.¼ 1.3 years). A total of 8214 (93.3%) students reported
activities (PIU-G) and those frequently engaged in Internet applica- normal Internet use, whereas 319 (3.6%) and 274 (3.1%) students met
tions other than gaming (PIU-NG). the criteria for PIU-G and PIU-NG, respectively. Although the adoles-
The Beck Depression Inventory II (BDI-II) (Beck et al., 1996) was cents classified as PIU-G were primarily male (27.0% female; 73.0%
used to assess depressive symptoms. The BDI-II contains 21 male), those categorized as PIU-NG were predominately female
questions with possible scores ranging from 0 to 3. Conduct (78.8% female; 21.2% male). The association between Internet use
disorder was assessed using the DISC Predictive Scales for conduct and gender was significant (χ²[2]¼ 166.0, po0.001).
disorder (DPS-CD) (Lucas et al., 2001), a diagnosis-specific self- Within the PIU-G group, 100% of the adolescents used the
report inventory. The total score ranges from 0 to 8 and corre- Internet for gaming; 97.5% used the Internet for chatting; 97.8%
sponds to the number of positive answers provided. The Strengths used the Internet for other Internet applications; and 45.9% used
and Difficulties Questionnaire (SDQ) (Goodman, 1997) was admi- the Internet for schoolwork. Within the PIU-NG group, 0% reported
nistered to assess various emotional and behavioral problems. The engaging in gaming; instead, 95.6% used the Internet for chatting;
present study used the sub-scales emotional symptoms, hyper- 96.4% used the Internet for other Internet applications; and 47.8%
activity/inattention, and peer problems. Each sub-scale contained used the Internet for schoolwork. Among students who showed
5 items, and the scores ranged from 0 to 10. normal Internet use, 35.3% used the Internet for gaming; 92.9%
The World Health Organization (WHO)-Five Well-being Index used it for chatting; 92.5% used it for other Internet applications;
(WHO-5) was used to measure dimensions of general psychologi- and 55.4% used it for schoolwork.
cal well-being (Primack, 2003). The raw score ranges from 0 to 25, The distributions of the dimensional scores (i.e., those regarding
where 0 represents the worst possible quality of life and 25 depression, conduct disorder, emotional symptoms, hyperactivity/
represents the best possible quality of life. For statistical analysis, inattention, peer problems, and well-being) for the three Internet
the WHO-5 raw score was converted into a percentage (0–100%). user groups are shown in Fig. 1. The frequencies of suicidal and self-
Suicidal behavior over the last year was assessed using the Paykel injurious behaviors for the three Internet groups are presented in
Suicide Scale (PSS) (Paykel et al., 1974), which includes all stages of Fig. 2. Table 2 presents detailed information on these variables for
suicidal tendency. Suicidal behavior was identified when any suicidal each group including the respective group differences. Pairwise
thoughts, suicide plans or suicide attempts were reported. A mod- group comparison of the different Internet use groups revealed
ified version (Brunner et al., 2014) of the Deliberate Self-Harm significant differences between the normal Internet use group and
Inventory (DSHI) (Gratz, 2001) was used to distinguish between PIU-G for BDI-II, DPS-CD, SDQ emotional symptoms, SDQ hyperac-
different forms of self-injurious behavior (e.g., cutting, burning, tivity, SDQ peer problems, PSS and DSHI. Likewise, significant
scratching, wound manipulation, and head banging). Self-injurious differences were found between the normal Internet use group
behavior was assessed when students reported any engagement in and PIU-NG regarding all psychological variables (BDI-II, DPS-CD,
these acts over the last year. SDQ emotional symptoms, SDQ hyperactivity, SDQ peer problems,
 E. Strittmatter et al. / Psychiatry Research 228 (2015) 128–135 131

Fig. 1. The distribution of abnormal psychology dimensional scores (i.e., depression, conduct disorder, reduced well-being, emotional symptoms, hyperactivity, and peer
problems) for the 3 Internet groups (normal Internet use, PIU-NG, and PIU-G). The boxplots show medians, first quartiles, third quartiles and outliers.

Fig. 2. The percentages (%) and their 95% confidence intervals of suicidal and self-injurious behaviors within the 3 Internet groups (normal Internet use, PIU-NG, and PIU-G).

WHO-5, PSS and DSHI) In the univariate approach comparison classified as PIU-G. The regression model explained 14.4% of the
between the groups of PIU-G and PIU-NG indicated significant information of the PIU categorization (Pseudo R2).
differences with respect to self-injurious behavior (p¼ 0.033), Post hoc comparisons of the relative risk ratios (RRR) confirmed
depression (po0.001), emotional symptoms (po0.001), well- some significant differences between adolescents with PIU-G and
being (po0.001) and peer problems (p¼0.028). those with PIU-NG with regards to gender (males: PIU-G4 PIU-
The results of multinomial logistic regression showed that NG, p o0.001), depression (PIU-NG4PIU-G, p ¼0.021) and peer
depression, conduct disorder, hyperactivity/inattention, and suici- problems according to SDQ (PIU-G 4PIU-NG, p ¼0.002). No sig-
dal and self-injurious behaviors were all associated with increased nificant differences were found between PIU-G and PIU-NG with
risks for both PIU-G and PIU-NG in relation to the group of normal regard to all other variables (emotional symptoms, conduct
Internet users (see Table 3). Males with reduced well-being, disorder, hyperactivity, well-being, suicidal and self-injurious
emotional symptoms and peer problems were more likely to be behaviors).
132 E. Strittmatter et al. / Psychiatry Research 228 (2015) 128–135

Table 2
Means and standard deviations (S.D.) of the dimensional variables, frequencies (n) and percentages (%) of the categorical variables for the three Internet user groups, and
effect sizes between groups (Cohen's d for continuous variables and odds ratios for categorical variables).

Normal PIU-G PIU-NG PIU-G vs. normal PIU-NG vs. normal PIU-NG vs. PIU-G

Mean S.D. Mean S.D. Mean S.D. Cohen's d Cohen's d Cohen's d

nnn nnn
Depression (BDI total score) 6.90 7.37 11.27 9.96 16.25 10.95 0.58 1.24 0.48nnn
Conduct disorder (DPS) 0.63 1.06 1.40 1.60 1.32 1.37 0.71nnn 0.64nnn  0.06
Emotional symptoms (SDQ) 2.70 2.14 3.43 2.41 4.36 2.38 0.34nnn 0.77nnn 0.39nnn
Hyperactivity (SDQ) 3.48 2.10 4.58 2.00 4.95 2.00 0.53nnn 0.70nnn 0.18
Peer problems (SDQ) 1.97 1.60 2.88 1.85 2.54 1.75 0.56nnn 0.35nnn  0.19n
Well-being (WHO-5) 61.22 18.67 59.22 19.05 50.68 20.11  0.11 -0.56nnn -0.44nnn
n % n % n % OR OR OR

Suicidal behaviors (PSS) 1748 21.36 144 45.86 146 53.48 3.12nnn 4.23nnn 1.36
Self-injurious behaviors (DSHI) 1450 18.05 119 38.51 131 49.06 2.84nnn 4.37nnn 1.54n

n
p o 0.05
nnn
p o0.001

Table 3
Multinomial logistic regression of gender, well-being, psychopathology, and self-harming behaviors among adolescents with PIU-G and those with PIU-NG.

Psychosocial variables PIU-G PIU-NG

RRR 95% CIs p RRR 95% CIs p

Male 4.30 3.24–5.71 o 0.001 0.45 0.32–0.62 o 0.001

Depressiona (BDI-II) 1.25 1.10–1.43 0.001 1.54 1.35–1.75 o 0.001
Conduct disordera (DPS) 1.24 1.14–1.36 o 0.001 1.30 1.18–1.45 o 0.001
Emotional symptomsa (SDQ) 1.20 1.04–1.38 0.014 1.04 0.89–1.21 0.637
Hyperactivitya (SDQ) 1.28 1.13–1.45 o 0.001 1.35 1.20–1.54 o 0.001
Peer problemsa (SDQ) 1.27 1.14–1.42 o 0.001 0.99 0.88–1.13 0.937
Well-beinga (WHO-5) 1.25 1.09–1.42 0.001 1.04 0.90–1.20 0.587
Suicidal behaviors (PSS) 1.82 1.37–2.42 o 0.001 1.36 1.01–1.84 0.045
Self-injurious behaviors (DSHI) 1.56 1.19–2.04 0.001 1.82 1.37–2.42 o 0.001

The regression coefficients are presented in their exponential form as relative risk ratios, with normal Internet use as base category. RRR: relative risk ratio; CIs: confidence
intervals; PIU-G: pathological Internet use-gaming; PIU-NG: pathological Internet use-non-gaming.
a
Standardized scores

4. Discussion (Aboujaoude et al., 2006) to 26.7% (Yu and Shek, 2013). These
wide ranges are probably due to differences among the countries,
This study investigated whether students with PIU-G and PIU- concepts, definitions and study samples used. The rates in the
NG have an elevated risk for psychopathology in comparison to present study (3.62% for PIU-G and 3.11% for PIU-NG) match those
normal Internet users. However, the main scope of the study was of other methodologically sound studies. Rates from 3% to 8.5% for
to address the impact of non-gaming Internet activities among online game addiction (Gentile et al., 2011; Van Rooij et al., 2011)
students with PIU, which (until now) has been scarcely investi- and from 4% to 4.4% for PIU (Durkee et al., 2012; Rumpf et al.,
gated. Within the PIU group, adolescents with PIU-G were cate- 2014) have been reported among adolescents in Europe. Our rates
gorized separately from those with PIU-NG. Furthermore, these also suggest that PIU-G and PIU-NG are equally distributed among
students were assessed with regard to their psychopathologies, students who have PIU; furthermore, a large proportion of
self-harming behaviors and psychological well-being. In summary, students with PIU might go undiagnosed using a classification
students with PIU-G and those with PIU-NG showed elevated system that refers exclusively to gaming.
levels of psychopathology and more self-harming behaviors com- The male gender was a strong and significant risk factor for
pared with normal Internet users. However, the differences PIU-G (RRR ¼4.3) but protective for PIU-NG (RRR ¼0.45). These
between students classified as PIU-G and those categorized as results are in line with those previously obtained by other authors
PIU-NG were marginal. who clearly separated PIU-G from PIU-NG (Mößle and Rehbein,
PIU was assessed using the YDQ, which was developed in line 2013). The present study clearly showed a significant increased
with the DSM-IV definition of pathological gaming (Young, 1998); risk for depression, conduct disorder, hyperactivity/inattention,
in fact, the YDQ shows strong similarities to the criteria of Internet suicidal and self-injurious behaviors for both PIU-groups in rela-
gaming disorder in the DSM-5 (see Table 1). However, whereas the tion to the group of normal Internet users.
DSM-5 only refers to online gaming, the concept of PIU as Thus far, most research has either investigated PIU in general or
measured by the YDQ includes other Internet applications (e.g., specifically online gaming in comparison to normal Internet use.
social Internet use or sexually oriented websites). Therefore, the Investigations of PIU have revealed associations with experience
frequency of Internet application use was assessed in combination of childhood abuse and insecure attachment (Schimmenti et al.,
with the YDQ, which enabled us to categorize students with PIU 2014b), depression (Ha et al., 2007; Yen et al., 2008), attention-
based on their gaming/non-gaming online activities. deficit/hyperactivity disorder (Ko et al., 2009), emotional problems
The reported prevalence of Internet gaming disorders ranges (Bakken et al., 2009), deliberate self-harming and suicidal behaviors
between 0.2% (Festl et al., 2013) and 46% (Wan and Chiou, 2006). (Kaess et al., 2014) and problems with peer relationships (Durkee
The prevalence of PIU shows a similar range, from 0.3% et al., 2012). The results concerning the relationship between PIU
 E. Strittmatter et al. / Psychiatry Research 228 (2015) 128–135 133

and well-being have been inconsistent (Caplan, 2002). Previous recently published cluster analysis differentiating the heterogeneity
studies have examined the social and psychological benefits of of psychological motives behind problematic online gaming and
Internet applications (Guan and Subrahmanyam, 2009); less calling for the development of personalized psychotherapeutic
research, however, has been performed on pathological Internet interventions (Billieux et al., 2015).
gaming. The studies conducted thus far on Internet gaming have Because of the cross-sectional nature of the present study and
similarly highlighted the associations between PIU and lower well- some of the inconsistent results among the few longitudinal
being (Festl et al., 2013; Lemmens et al., 2011), emotional problems studies available, future prospective research is needed to deter-
(Gentile et al., 2011), attention-deficit/hyperactivity disorder mine directional and causal relationships among PIU (gaming and
(Gentile, 2009), depression (Van Rooij et al., 2011) and suicidal non-gaming), comorbid psychopathologies, self-harm and well-
ideation (Wenzel et al., 2009). Based on previous data, it is possible being.
that most of the psychological burden associated with addictive
Internet behaviors are caused by Internet gaming. However, our 4.1. Study strengths and weaknesses
data clearly indicate that PIU can occur without engagement in
online gaming activities, and these behaviors might have similar The present study was conducted using a large representative
negative mental health consequences. sample of adolescents across different European countries. How-
To our knowledge, this study is the first to directly compare PIU-G ever, the study design was cross-sectional, and the information
and PIU-NG using the same assessment methodology in a large, was based on self-report questionnaires rather than structured
multi-site European sample. Our results showed that both PIU-G and clinical interviews. Although the YDQ has sometimes been criti-
PIU-NG are associated with significant and elevated risks for psy- cized in the past, the comparison between the YDQ and the DSM-5
chopathologies and self-harming behaviors when compared to criteria for Internet gaming disorder shows a large similarity with
normal Internet users. The findings are critical for both prevention regard to both the criteria and the cut-off. As previous research has
and treatment. As PIU-NG is associated with a significant increased already highlightened that sub-threshold problematic Internet use
risk for impairment and abnormal psychology (in comparison to not only has the same risk factors as PIU but also is associated with
adolescents without PIU), personalized interventions should be significant psychopathology (Durkee et al., 2012; Kaess et al.,
developed and offered. However, if Internet gaming disorder even- 2014), the present study did not further differentiate an at-risk
tually becomes an official diagnosis, based on these findings 3.1% of Internet use group. In contrast, the group of normal Internet users
European adolescents who suffer from PIU-NG would not be eligible includes the group of problematic or at-risk Internet users.
for appropriate mental health consultation or therapy. The current However, even though the sub-threshold group is included in
DSM-5 criteria will likely narrow research activities to focus on the non-PIU group, our results highlight significantly elevated
online gaming only. In addition, additional reasons might exist as to relative risk ratios for most psychopathologies, which underline
why PIU-NG should be considered in future diagnostic criteria. For the strong effects. To further differentiate the pathological non-
example, overlap exists between general PIU and pathological gam- gaming Internet activities would have been very interesting. As
ing with many contemporary Internet games based on the concept this comparison is beyond the scope of our paper, future research
that social interactions are necessary to solve problems. Last but not should address this important topic. To our knowledge, our study
least, our study showed that individuals with PIU-G also used other is the first to compare groups of adolescents with PIU-G or PIU-
Internet applications to a large extent. In sum, the differences among NG; therefore, it adds to the current knowledge and offers a
online activities might share a variety of addictive characteristics that valuable contribution to the problem of defining the features of
could be hard to distinguish because engagement in diverse online this condition (Petry and O’Brien, 2013).
activities is likely the norm among adolescents. In fact, these findings
should encourage questioning the recent decision of the APA to 4.2. Conclusion
include only Internet gaming within the new diagnostic concept
regarding the pathological and addictive developments of Internet Internet gaming disorder was included in Section 3 of the DSM-5
behaviors (APA 2013) and might provide support for the broader for the first time. Although other Internet applications are not
inclusion of different online activities within the concept of “com- currently mentioned in the DSM-5 (Petry and O’Brien, 2013), some
puter and Internet addiction” in the upcoming ICD-11. authors have suggested that they have addictive potential as well
Regarding the groups of PIU-G and PIU-NG, some differences (Caplan, 2002; Meerkerk et al., 2006; van den Eijnden et al., 2008).
were found: the students in the PIU-NG group were predomi- The present study clearly demonstrated that adolescents with PIU-G
nantly female and had higher amounts of depression. Because girls and those with PIU-NG are generally similar with regard to their
are more likely to be depressed, the possibility of a gender effect related psychopathologies and self-harming behaviors. Thus, our
exists. However, gender was included in the multinomial logistic data add to the body of evidence showing that PIU-NG should be
regression so any potential “gender effect” was already controlled. considered in the future diagnostic criteria for Internet-related
Individuals with PIU-NG might use the Internet as a way to self- addictive behaviors.
medicate; thus, conscientious people who seek social contact and
receive support in times of social withdrawal might not have a
reduction in well-being. Contributors
Students in the PIU-G group were predominantly male and
showed more peer problems. On the one hand, excessive gaming DW was the project leader and coordinator of the WE-STAY
might lead to difficulties in engaging with people in real life and the project. VC, CWH, CW, and MS were members of the executive
deprivation of face-to-face contacts. On the other hand, excessive committee. MS, AA, JB, RB, DC and DW were site leaders of the
gamers might feel less competent in social relationships, and their study and supervised the design and implementation at the
virtual lives might offer the chance to commit themselves to respective study sites. GF, RB, MS and PV were involved in the
relationships within the confident interspace needed to present data collection for the study. ES and MK made a significant
their ideal selves. Students with PIU might prefer to hide behind contribution to the analysis and interpretation of data. PP partici-
avatars. Apart from this possibility, gaming might be a way to escape pated in the design of the study and performed the statistical
the complex nature of the conflicts of every-day life and achieve analysis. ES drafted the manuscript. MK supervised the data
rewards more easily. These considerations are strengthened by a analysis and writing of the manuscript including critical revision
134 E. Strittmatter et al. / Psychiatry Research 228 (2015) 128–135

of the manuscript. PP revised the article critically. All authors Carli, V., Wasserman, C., Wasserman, D., Sarchiapone, M., Apter, A., Balazs, J., Bobes, J.,
reviewed the manuscript for intellectual content. All authors read Brunner, R., Corocan, P., Cosman, D., Guillemin, F., Haring, C., Kaess, M., Kahn, J.P.,
Keeley, H., Kereszteny, A., Iosue, M., Mars, U., Musa, G., Nemes, B., Postuvan, V.,
and approved the final version of the manuscript. Reiter-Theil, S., Saiz, P., Varnik, P., Varnik, A., Hoven, C.W., 2013. The saving and
empowering young lives in Europe (SEYLE) randomized controlled trial (RCT):
methodological issues and participant characteristics. BMC Public Health 13, 479.
Durkee, T., Kaess, M., Carli, V., Parzer, P., Wasserman, C., Floderus, B., Apter, A.,
Conflict of interest Balazs, J., Barzilay, S., Bobes, J., Brunner, R., Corcoran, P., Cosman, D., Cotter, P.,
Despalins, R., Graber, N., Guillemin, F., Haring, C., Kahn, J.P., Mandelli, L.,
The authors declare that they have no conflict of interest. Marusic, D., Meszaros, G., Musa, G.J., Postuvan, V., Resch, F., Saiz, P.A., Sisak,
M., Varnik, A., Sarchiapone, M., Hoven, C.W., Wasserman, D., 2012. Prevalence of
pathological Internet use among adolescents in Europe: demographic and
social factors. Addiction 107 (12), 2210–2222.
Acknowledgment Festl, R., Scharkow, M., Quandt, T., 2013. Problematic computer game use among
adolescents, younger and older adults. Addiction 108 (3), 592–599.
Fu, K., Chan, W.S.C., Wong, P.W.C., Yip, P.S.F., 2010. Internet addiction: prevalence,
The Coordination Theme 1 (Health) of the European Union discriminant validity and correlates among adolescents in Hong Kong. The
Seventh Framework Program (FP7), Grant agreement nr HEALTH- British Journal of Psychiatry 196 (6), 486–492.
Gentile, D., 2009. Pathological video-game use among youth ages 8 to 18: a national
2010–241542, supports the Working in Europe to Stop Truancy study. Psychological Science 20 (5), 594–602.
Among Youth (WE-STAY) project. The authors were independent Gentile, D.A., Choo, H., Liau, A., Sim, T., Li, D., Fung, D., Khoo, A., 2011. Pathological
of the funders with regard to all aspects of the study design, data video game use among youths: a two-year longitudinal study. Pediatrics 127 (2),
analysis, and writing of this manuscript. The Project Leader and e319–e329.
Goodman, R., 1997. The strengths and difficulties questionnaire: a research note.
Coordinator of the WE-STAY project is Professor of Psychiatry and Journal of Child Psychology and Psychiatry, and Allied Disciplines 38 (5),
Suicidology Danuta Wasserman at Karolinska Institute (KI) and 581–586.
Head of the National Center for Suicide Research and the Preven- Gratz, K., 2001. Measurement of Deliberate self-harm: preliminary data on the
deliberate self-harm inventory. Journal of Psychopathology and Behavioral
tion of Mental Illness and Suicide (NASP) at KI, Stockholm, Assessment 23 (4), 253–263.
Sweden. Other members of the executive committee include Guan, S.-S.A., Subrahmanyam, K., 2009. Youth Internet use: risks and opportunities.
Professor Marco Sarchiapone, Department of Health Sciences, Current Opinion in Psychiatry 22 (4), 351–356.
Ha, J.H., Kim, S.Y., Bae, S.C., Bae, S., Kim, H., Sim, M., Lyoo, I.K., Cho, S.C., 2007.
University of Molise, Campobasso, Italy; Senior Lecturer Vladimir
Depression and Internet addiction in adolescents. Psychopathology 40 (6),
Carli, NASP at KI, Stockholm, Sweden; and Professor Christina W. 424–430.
Hoven and Anthropologist Camilla Wasserman, Department of Kaess, M., Durkee, T., Brunner, R., Carli, V., Parzer, P., Wasserman, C., Sarciapone, M.,
Child and Adolescent Psychiatry, Columbia University- New York Hoven, C., Apter, A., Balazs, J., Balint, M., Bobes, J., Cohen, R., Cosman, D., Cotter,
P., Fischer, G., Floderus, B., Iosue, M., Haring, C., Kahn, J.P., Musa, G.J., Nemes, B.,
State Psychiatric Institute, New York, USA. The WE-STAY Consor- Postuvan, V., Resch, F., Saiz, P.A., Sisask, M., Snir, A., Varnik, A., Ziberna, J.,
tium includes centers located in 10 European countries. The site Wasserman, D., 2014. Pathological Internet use among European adolescents:
leaders for each respective center and country are Danuta Wasser- psychopathology and self-destructive behaviours. European Child & Adolescent
Psychiatry 23 (11), 1093–1102.
man (NASP at KI, Sweden, Coordinating Center), Airi Varnik
Kim, M.G., Kim, J., 2010. Cross-validation of reliability, convergent and discriminant
(Estonian Swedish Mental Health & Suicidology Institute, Estonia), validity for the problematic online game use scale. Computers in Human
Jean-Pierre Kahn (University of Nancy, France), Romuald Brunner Behavior 26 (3), 389–398.
(University of Heidelberg, Germany), Judit Balazs (Vadaskert Child King, D.L., Haagsma, M.C., Delfabbro, P.H., Gradisar, M., Griffiths, M.D., 2013. Toward
a consensus definition of pathological video-gaming: a systematic review of
and Adolescent Psychiatric Hospital, Hungary), Alan Apter (Schnei- psychometric assessment tools. Clinical Psychology Review 33 (3), 331–342.
der Children's Medical Center of Israel, Tel-Aviv University, Tel Ko, C.-H., Yen, J.-Y., Chen, C.-C., Chen, S.-H., Yen, C.-F., 2005. Gender differences and
Aviv, Israel), Marco Sarchiapone (University of Molise, Italy), Doina related factors affecting online gaming addiction among Taiwanese adoles-
cents. The Journal of Nervous and Mental Disease 193 (4), 273–277.
Cosman (Iuliu Hatieganu University of Medicine and Pharmacy, Ko, C.-H., Yen, J.-Y., Liu, S.-C., Huang, C.-F., Yen, C.-F., 2009. The associations between
Romania), Vita Postuvan (University of Primorska, Slovenia) and aggressive behaviors and Internet addiction and online activities in adoles-
Julio Bobes (University of Oviedo, Spain). The Medical Faculty of cents. The Journal of Adolescent Health 44 (6), 598–605.
Lemmens, J.S., Valkenburg, P.M., Peter, J., 2011. Psychosocial causes and conse-
the University of Heidelberg supported the analyses.
quences of pathological gaming. Computers in Human Behavior 27 (1),
144–152.
Lucas, C.P., Zhang, H., Fisher, P.W., Shaffer, D., Regier, D.A., Narrow, W.E., Bourdon,
References K., Dulcan, M.K., Canino, G., Rubio-Stipec, M., Lahey, B.B., Friman, P., 2001. The
DISC Predictive Scales (DPS): efficiently screening for diagnoses. Journal of the
Aboujaoude, E., Koran, L.M., Gamel, N., Large, M.D., Serpe, R.T., 2006. Potential American Academy of Child and Adolescent Psychiatry 40 (4), 443–449.
markers for problematic Internet use: a telephone survey of 2513 adults. CNS Mann, K., Fauth-Bühler, M., Seiferth, N., Heinz, A., Expertengruppe Verhaltens-
Spectrums 11 (10), 750–755. süchte der, D.G.P.P.N., 2013. The concept of behavioral addiction and limits of
Bakken, I.J., Wenzel, H.G., Götestam, K.G., Johansson, A., Oren, A., 2009. Internet the term addiction. Der Nervenarzt 84 (5), 548–556.
addiction among Norwegian adults: a stratified probability sample study. Meerkerk, G.-J., Van Den Eijnden, R.J.J.M., Garretsen, H.F.L., 2006. Predicting
Scandinavian Journal of Psychology 50 (2), 121–127. compulsive Internet use: it’s all about sex! cyberpsychology & behavior: the
Beck, A.T., Steer, R.A., Ball, R., Ranieri, W., 1996. Comparison of Beck depression impact of the Internet. Multimedia and Virtual Reality on Behavior and Society
inventories -IA and -II in psychiatric outpatients. Journal of Personality 9 (1), 95–103.
Assessment 67 (3), 588–597. Mößle, T., Rehbein, F., 2013. Predictors of problematic video game usage in
Billieux, J., Thorens, G., Khazaal, Y., Zullino, D., Achab, S., van der Linden, M., 2015. childhood and adolescence. SUCHT - Zeitschrift Für Wissenschaft Und Praxis/
Problematic involvement in online games: a cluster analytic approach. Com- Journal of Addiction Research and Practice 59 (3), 153–164.
puters in Human Behavior 43, 242–250. Paykel, E.S., Myers, J.K., Lindenthal, J.J., Tanner, J., 1974. Suicidal feelings in the
Brunner, R., Kaess, M., Parzer, P., Fischer, G., Carli, V., Hoven, C.W., Wasserman, C., general population: a prevalence study. The British Journal of Psychiatry 124
Sarciapone, M., Resch, F., Apter, A., Balazs, J., Barzilay, S., Bobes, J., Corcoran, P., (0), 460–469.
Cosman, D., Haring, C., Iosuec, M., Kahn, J.P., Keeley, H., Meszaros, G., Nemes, B., Petry, N.M., O’Brien, C.P., 2013. Internet gaming disorder and the DSM-5. Addiction
Podlogar, T., Postuvan, V., Saiz, P.A., Sisak, M., Tubiana, A., Varnik, A., Wasser- 108 (7), 1186–1187.
man, D., 2014. Life-time prevalence and psychosocial correlates of adolescent Petry, N.M., Rehbein, F., Gentile, D.A., Lemmens, J.S., Rumpf, H.-J., Mößle, T., Bischof,
direct self-injurious behavior: a comparative study of findings in 11 European G., Tao, R., Fung, D.S., Borges, G., Auriacombe, M., Gonzalez Ibanez, A., Tam, P.,
countries. Journal of Child Psychology and Psychiatry, and Allied Disciplines 55 O’Brien, C.P., 2014. An international consensus for assessing Internet gaming
(4), 337–348. disorder using the new DSM-5 approach. Addiction 109 (9), 1399–1406.
Cao, F., Su, L., Liu, T., Gao, X., 2007. The relationship between impulsivity and Primack, B.A., 2003. The WHO-5 wellbeing index performed the best in screening
Internet addiction in a sample of Chinese adolescents. European Psychiatry 22 for depression in primary care. ACP Journal Club 139 (2), 48.
(7), 466–471. Rumpf, H.-J., Vermulst, A.A., Bischof, A., Kastirke, N., Gürtler, D., Bischof, G.,
Caplan, S.E., 2002. Problematic Internet use and psychosocial well-being: develop- Meerkerk, G.J., John, U., Meyer, C., 2014. Occurence of Internet addiction in a
ment of a theory-based cognitive–behavioral measurement instrument. Com- general population sample: a latent class analysis. European Addiction
puters in Human Behavior 18 (5), 553–575. Research 20 (4), 159–166.
 E. Strittmatter et al. / Psychiatry Research 228 (2015) 128–135 135

Schimmenti, A., Caretti, V., La Barbera, D., 2014a. Internet gaming disorder or Internet Postuvan, C.W., Reiter-Theil, S., Resch, F., Saiz, P., Sarchiapone, M., Sisask, M., Varnik,
addiction? A plea for conceptual clarity. Clinical Neuropsychiatry 11 (5), 145–146. A., Hoven, C.W., 2010. Saving and empowering young lives in Europe (SEYLE): a
Schimmenti, A., Passanisi, A., Gervasi, A.M., Manzella, S., Fama, F.I., 2014b. Insecure randomized controlled trial. BMC Public Health 10, 192.
attachment attitudes in the onset of problematic Internet use among late Wenzel, H.G., Bakken, I.J., Johansson, A., Götestam, K.G., Øren, A., 2009. Excessive
adolescents. Child Psychiatry and Human Development 45, 588–595. computer game playing among Norwegian adults: self-reported consequences
Van den Eijnden, R.J.J.M., Meerkerk, G.-J., Vermulst, A.A., Spijkerman, R., Engels, R.C. of playing and association with mental health problems. Psychological Reports
M.E., 2008. Online communication, compulsive Internet use, and psychosocial 105 (3 Pt 2), 1237–1247.
well-being among adolescents: a longitudinal study. Developmental Psychol- Wieland, D.M., 2005. Computer addiction: implications for nursing psychotherapy
ogy 44 (3), 655–665. practice. Perspectives in Psychiatric Care 41 (4), 153–161.
Van Rooij, A.J., Schoenmakers, T.M., Vermulst, A.A., Van den Eijnden, R.J.J.M., Van de Yen, J.-Y., Ko, C.-H., Yen, C.-F., Chen, S.-H., Chung, W.-L., Chen, C.-C., 2008. Psychiatric
Mheen, D., 2011. Online video game addiction: identification of addicted
symptoms in adolescents with Internet addiction: comparison with substance
adolescent gamers. Addiction 106 (1), 205–212.
use. Psychiatry and Clinical Neurosciences 62 (1), 9–16.
Wan, C.-S., Chiou, W.-B., 2006. Psychological motives and online games addiction: a
Young, K.S., 1998. Caught in the net: how to recognize the signs of Internet
test of flow theory and humanistic needs theory for Taiwanese adolescents.
addiction-and a winning strategy for recovery. J. Wiley, New York.
Cyberpsychology & Behavior 9 (3), 317–324.
Yu, L., Shek, D.T.L., 2013. Internet addiction in Hong Kong adolescents: a three-year
Wasserman, D., Carli, V., Wasserman, C., Apter, A., Balazs, J., Bobes, J., Bracale, R., Brunner,
R., Bursztein-Lipsicas, C., Corcoran, P., Cosman, D., Durkee, T., Feldman, D., Gadoros, J., longitudinal study. Journal of Pediatric and Adolescent Gynecology 26 (3 Suppl),
Guillemin, F., Haring, C., Kahn, J.P., Kaess, M., Keeley, H., Marusic, D., Nemes, B., S10–S17.