Content uploaded by Azharul Islam
Author content
All content in this area was uploaded by Azharul Islam on Dec 01, 2016
Content may be subject to copyright.
Available via license: CC BY 4.0
Content may be subject to copyright.
Prevalence andrisk factors
ofproblematic internet use andthe associated
psychological distress amonggraduate students
ofBangladesh
Md. Azharul Islam1* and Muhammad Zakir Hossin2
Background
Problematic Internet use (PIU) or Internet addiction (IA) has been extensively
researched since the mid-1990s, particularly in some Western and Asian countries.
Although considerable evidence shows that PIU is associated with a number of negative
health outcomes in adolescents and adults (Ko etal. 2012; Kuss etal. 2013a), it was not
officially classified as a clinical disorder in the latest edition of the Diagnostic and Sta-
tistical Manual for Mental Disorders (DSM-V) (APA 2013). is indicates the need for
further evidence on this emerging mental health epidemic.
Examining the prevalence and factors of PIU and the associated health problems is
particularly important in a country like Bangladesh where the growth of Internet use is
faster than socio-economic development itself. In line with global technological devel-
opment, the use of Internet has also accelerated in Bangladesh, with the number of
Internet users increasing from 0.1 million in 2000 to 62 million in 2016 (BTRC 2016).
Abstract
A growing body of epidemiological literature suggests that problematic Internet use
(PIU) is associated with a range of psychological health problems in adolescents and
young adults. This study aimed to explore socio-demographic and behavioural cor-
relates of PIU and examine its association with psychological distress. A total of 573
graduate students from Dhaka University of Bangladesh responded to a self-adminis-
tered questionnaire that included internet addiction test (IAT), 12-items General Health
Questionnaire and a set of socio-demographic and behavioural factors. The study
found that nearly 24% of the participants displayed PIU on the IAT scale. The preva-
lence of PIU significantly varied depending on gender, socioeconomic status, smoking
habit and physical activity (p < 0.05). The multiple regression analyses suggested that
PIU is strongly associated with psychological distress regardless of all other explanatory
variables (adjusted OR 2.37, 95% CI 1.57, 3.58). Further research is warranted to confirm
this association by employing prospective study designs.
Keywords: Internet addiction test, General health questionnaires, Socioeconomic
status, Young adults, Bangladesh
Open Access
© The Author(s) 2016. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License
(http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium,
provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and
indicate if changes were made.
RESEARCH ARTICLE
Islam and Hossin
Asian J of Gambling Issues and Public Health (2016) 6:11
DOI 10.1186/s40405-016-0020-1
*Correspondence:
azharulislam@du.ac.bd
1 Department of Educational
and Counselling Psychology,
University of Dhaka,
Dhaka 1000, Bangladesh
Full list of author information
is available at the end of the
article
Page 2 of 14
Islam and Hossin Asian J of Gambling Issues and Public Health (2016) 6:11
is unprecedented growth of Internet may pose a variety of health challenges, if proper
measures are not taken in the right time. Perhaps the most important reason is the
unconditional and positive acceptance of its use among common users as well as policy
makers. e proliferation of the use of Internet has been largely considered an indica-
tor of development. e government gives tremendous effort to make this happen, that
is, to provide each citizen access to Internet with the ultimate aim of building a ‘Digital
Bangladesh’ by 2021. erefore, it is plausible to assume that people as a whole are not
aware of the problematic use of Internet and its effects on their psychological health.
PIU has been defined as ‘an individual’s inability to control their internet use, which
in turn leads to feelings of distress and functional impairment of daily activities’ (Sha-
pira et al. 2000). It is a condition akin to ‘impulse-control disorder’ that does not
involve an intoxicant similar to symptoms of pathological gambling, overeating and so
on (Young 2004). Kimberly S. Young was one of the key proponents of the construct
‘Internet Addiction’. She proposed eight diagnostic criteria adapted from the symptoms
of pathological gambling (DSM-IV) for detecting an individual’s Internet use as prob-
lematic (Young 1996). ese are cognitive preoccupation with the Internet, increased
tolerance, unsuccessful attempts to decrease use, withdrawal symptoms, staying online
much longer than needed, lying about online activity, negative emotion resulting from
online activity, and use of the Internet to self-medicate. Clearly, an individual with these
symptoms will face difficulty in normal daily functioning and may experience poor men-
tal health. To test this hypothesis, it is crucial for professionals to measure PIU system-
atically. A number of tools have been developed out of which Young’s internet addiction
test (IAT) was found to have wide uses (Young 1998). IAT gives a score ranging between
20 and 100, with a higher score indicating a greater degree of PIU.
Studies examining the prevalence of PIU lack consensus on the diagnostic criteria
set to identify PIU due to various measures and cut off points used (Kuss etal. 2013a,
b). Considering only Young’s IAT with a cut-off point of 70 and above, a recent meta-
analysis of 164 prevalent figures from 31 nations across seven world regions estimated
a global prevalence of IA as 6.0% (95% CI 5.1–6.9) (Cheng and Li 2014). is classifica-
tion of PIU, however, did not consider moderate users who are also unable to control
their Internet use (Young 2016). erefore, incorporating moderate and excessive users
as defined by IAT would be safe to detect PIU (i.e., PIU=IAT score 50+). Following
this notion, the prevalence of PIU appears to be high. For instance, a recent epidemio-
logical study conducted in six Asian countries revealed the prevalence of PIU as follows:
Philippines (51%), Japan (48%), China (19%), Hong Kong (35%), South Korea (14%), and
Malaysia (37.5%) (Mak etal. 2014). No such comprehensive investigations have been
carried out in Bangladesh, though a preliminary assessment with a small sample of uni-
versity students (n=172) found that 36% of the participants scored equivalent to PIU
(Nigar and Karim 2014).
e majority of the studies, however, reported the prevalence, predictors and conse-
quences of addictive use of Internet using adolescent samples (e.g., Kuss etal. 2013b;
Lam etal. 2009; Liberatore etal. 2011; Yu and Shek 2013). Much less is known about
the young adults, notably the university graduates, who are at particular risk of PIU due
to several reasons. Firstly, they have comparatively wide and easy access to the Internet
via campus laboratory, free campus Wifi Zone and cheap mobile Internet packages. is
Page 3 of 14
Islam and Hossin Asian J of Gambling Issues and Public Health (2016) 6:11
explains, in part, why 90% of the social media users in the USA belong to the age group
18–29 (Perrin 2015). In Bangladesh, 94% of the total subscribers accessed Internet via
mobile phones until April 2016 (BTRC 2016) and obviously, youths are the major con-
sumers of the mobile Internets. Secondly, the youth, irrespective of their living arrange-
ment such as living with family or in university dormitories, typically enjoy the freedom
of their choices and Internet use. Next, considering their developmental period in which
they are striving to build own identity, career and partner, they may use Internet to facil-
itate the expected growth, which in turn may appear as an addictive behaviour (Lanthier
and Windham 2004). According to Wallace, ‘there is no question that twernty-first cen-
tury youth have become far more dependent upon connectivity for studying, playing,
communicating, and socializing’ (Wallace 2014). Existing literature also demonstrates a
prevalence of PIU among university/college students from 0.8% in Italy (Poli and Agrimi
2012) to a moderate level of 18.3% in UK (Niemz etal. 2005) to a higher level of 40% in
Jordan (Al-Gamal etal. 2015).
PIU has been found to be associated with a broad range of negative health outcomes
with some exceptions. For example, increased depressive symptoms were associated
with spending increased time in online activities namely shopping and gambling but
inversely with chatting, communication and email (Morgan and Cotten 2003). at is,
spending time on social networking was positively associated with mental health. How-
ever, recent evidence shows an overall negative relation with online networking and
well-being (Sabatini and Sarracino 2014) which indicates that the normal use of Internet
could turn into problematic over time. In addition to well-being, PIU was also linked
with adolescents’ psychosomatic (i.e., poor physical energy, physiological dysfunction,
and poorer immunity), behavioural and emotional symptoms (Cao etal. 2011). Adoles-
cents with PIU are also at risk of relatively high depression and poor social adaptation.
Lam and Peng (2010), for example, prospectively examined the effect of pathologi-
cal Internet use and demonstrated that the adolescents with addictive use of Internet
were at a relatively high risk of developing depression in the follow up. Although Inter-
net provides an avenue for easy social interactions, transferring these skills into real life
is difficult, as it requires exposures to real life situations. Furthermore, psychiatric co-
morbidity of IA has also been well-documented particularly for substance use disorder,
attention-deficit hyperactivity disorder (ADHD), depression, hostility, and social anxiety
disorder [see Ko etal. (2012) for a review].
Specific studies on PIU and psychological well-being among university students were
limited to explaining simple association of these two constructs. For instance, PIU was
positively correlated with psychological distress as measured by perceived stress scale
(Cohen etal. 1983) in a sample of university students in Jordan (Al-Gamal etal. 2015).
Niemz etal.(2005) assessed psychological distress by the 12-items General Health Ques-
tionnaire (GHQ-12) and PIU of 371 British students and found that there was no sig-
nificant difference in the mean score of GHQ-12 among PIU and non-PIU groups [F(2,
368)=2.15; p=0.118].
While there is a plethora of research focusing on a general negative association of PIU
with psychological well-being, very little is known about the role of the socio-demo-
graphic determinants or the lifestyle factors in the association of interest. For example,
socio-demographic factors such as age, sex, socio-economic status (SES), relationship
Page 4 of 14
Islam and Hossin Asian J of Gambling Issues and Public Health (2016) 6:11
status, living arrangements as well as behavioural factors such as smoking, physical
activity and sleeping habit are generally found to be associated with overall psychologi-
cal well-being (Bayram and Bilgel 2008; Glozier etal. 2010; Matsuzaki etal. 2007). Short
habitual sleep duration, for instance, has been shown to be linearly associated with per-
sistent psychological distress in young adults (Glozier etal. 2010). Again, a breakup in a
committed relationship usually results in symptoms of depression, loneliness and even
self-harm tendencies (Rhoades etal. 2011). Furthermore, the lower the SES, the poorer
the overall health, popularly known as the social gradient in health (Marmot 2009). In
a similar vein, males generally report a higher prevalence of PIU than females (Ha and
Hwang 2014; Shahnaz and Karim 2014). ese factors, in turn, could play a crucial role
in the association between PIU and psychological distress. erefore, an individual with
lower psychological well-being could be the victim of many other factors and not neces-
sarily due to his or her problematic use of Internet. Consequently, an investigation into
the association between PIU and psychological well-being with a simultaneous focus on
the relevant socio-demographic and lifestyle factors would be a useful contribution.
To our knowledge, only three studies with comparatively small sample sizes focused
on IA in Bangladesh (Karim and Nigar 2014; Nigar and Karim 2014; Shahnaz and
Karim 2014) but none investigated IA in relation to psychological distress or its socio-
demographic and behavioural determinants. e present paper, therefore, is the first of
its kind in the context of Bangladeshi society. e paper aims to report the prevalence
and risk factors of PIU among young Bangladeshi adults and examine the association
between PIU and psychological distress with a particular attention paid to the role of
socio-demographic and health behavioural factors in the association of interest. Specifi-
cally, the paper addresses the following research questions: (1) What is the prevalence
of PIU among the graduate students of Bangladesh? (2) Does the prevalence of PIU dif-
fer significantly between different social groups? (3) Is PIU associated with psychologi-
cal distress independent of the sociodemographic and behavioural factors? (4) To what
extent is the association between PIU and psychological distress explained by the socio-
demographic and behavioural factors?
Methods
Participants
e participants of this study were the graduate students of Dhaka University (DU) of
Bangladesh. DU is one of the oldest universities in the South Asian region, established
in 1921. Currently, there are around 37,000 students pursuing their higher studies in 77
different departments, categorized under 13 faculties and 11 institutes. Studying at DU
is almost free, as the state bears almost the entire tuition. erefore, students from all
socioeconomic backgrounds can pursue higher studies once they pass the very com-
petitive entrance exam. We chose DU for this study due to its diverse socio-economic
characteristics of the students. Around 600 graduates of DU from various faculties were
approached for data collection, during January and February 2015. Out of them, data
from 573 participants were kept for final analyses as a few of them refused to take part
and some returned the incomplete questionnaires. Participants were selected based on
a stratified random sampling design and the sample was representative of all faculties. A
detailed description of the study sample can be found in Table1.
Page 5 of 14
Islam and Hossin Asian J of Gambling Issues and Public Health (2016) 6:11
Procedure
A written informed consent was obtained from each participant before collecting the
data. Participation in the study was voluntary. Participants were not given any privilege
or not discriminated in any means due to their participation or not participation. Prior to
commencing the study, ethical approval was taken from the concerned university ethics
committee. Five research assistants holding at least a Master degree in psychology/clini-
cal/counselling psychology were employed for data collection. ey were given necessary
training on the research tool, data collection procedure and ethical issues of research.
Measures
Psychological distress
e main dependent variable in the study is psychological distress. e Bangla validated
(Sorcar and Rahman 1990) 12-item version of General Health Questionnaire (Goldberg
Table 1 Sample distribution and the prevalence of problematic Internet use by socio-
demographic andbehavioural factors
*Instead of p-for-dierence, p-for-trend has been reported for socioeconomic status
Characteristics Sample Problematic Internet use p-for-dierence
Yes No
% (n) % (n) % (n)
Total 100 (573) 23.9 (137) 76.1 (436)
Age (years) 0.520
20–25 60.7 (348) 23.0 (80) 77.0 (268)
26–30 39.3 (225) 25.3 (57) 74.7 (168)
Sex <0.001
Female 30.2 (173) 12.7 (22) 87.3 (151)
Male 69.8 (400) 28.8 (115) 71.2 (285)
Socioeconomic status <0.05*
High 22.7 (130) 19.2 (25) 80.8 (105)
Middle 62.0 (355) 22.8 (81) 77.2 (274)
Low 15.4 (88) 35.2 (31) 64.8 (57)
Relationship status 0.071
Single 42.6 (244) 24.6 (60) 75.4 (184)
Partnered 40.3 (231) 19.9 (46) 80.1 (185)
Separated 17.1 (98) 31.6 (31) 68.4 (67)
Living arrangement 0.930
With family 29.5 (169) 23.7 (40) 76.3 (129)
Dorm/mess 70.5 (404) 24.0 (97) 76.0 (307)
Smoking <0.05
No 80.1 (459) 21.8 (100) 78.2 (359)
Yes 19.9 (114) 32.5 (37) 67.5 (77)
Physical activity <0.05
No 49.4 (283) 27.9 (79) 72.1 (204)
Yes 50.6 (290) 20.0 (58) 80.0 (232)
Sleep duration 0.693
Normal sleep 66.1 (379) 23.0 (87) 77.0 (292)
Short sleep 29.7 (170) 25.3 (43) 74.7 (127)
Long sleep 4.2 (24) 29.2 (7) 70.8 (17)
Page 6 of 14
Islam and Hossin Asian J of Gambling Issues and Public Health (2016) 6:11
1972) (GHQ-12) was used to measure psychological distress. is is one of the widely
used self-reported measures of general psychological health. is instrument was origi-
nally developed as a screening test for detecting minor psychiatric disturbance or strain.
e measure assesses changes in affective and somatic symptoms relative to usual levels
of health, e.g., feelings of strain, depression, inability to cope, anxiety-based insomnia
and lack of confidence (Goodchild and Duncan-Jones 1985). ere are four different
scoring procedures available for GHQ-12 (Goldberg etal. 1998). In this study, we fol-
lowed the 0–0–1–1 scoring procedure which produces a score ranging from 0 to 12,
with a higher score indicating higher psychological distress. Again, there are various
thresholds for detecting psychological distressed. For a safe detection, the mean score
has been suggested as a potential threshold (Goldberg etal. 1998). e mean GHQ-12
score for this study was 3.03. erefore, we chose ‘three’ as the cut-off point for detect-
ing psychologically distressed cases. Internal consistency of the Bangla GHQ-12 for this
sample was very good (Chornbach’s α=0.85).
PIU
e main independent variable used in the study is PIU which was assessed by Young’s
IAT (Young 1996), the first psychometrically valid tool to measure problematic use of
Internet. is 20-item scale is designed to measure psychological dependence, compul-
sive use, and withdrawal as well as related problems of school, sleep, family, and time
management. Every item is scored on a five-point Likert scale ranging from 1 (rarely) to
5 (always). Total score is the sum of all items giving a range of 20 to 100 where a higher
score indicates greater level of IA. For this study the Bangla validated IAT (Karim and
Nigar 2014) was used. e Bangla version of IAT retained 18 items instead of original
20 items with a four factor structure. e score of Bangla IAT, therefore, ranges between
18 and 90. ere is no gold standard for distinguishing between PIU and non-PIU (Kuss
etal. 2013a, b) with cut points varying from 50 (Kormas etal. 2011) to 80 (Liberatore
etal. 2011; Ostovar etal. 2016). According to the IAT manual, users are given different
labels based on the total score obtained on the IAT scale. ese are: normal user (IAT
total ≤30), mild user (IAT total=31–49), moderate user (IAT total=50–79) and severe
or excessive user (IAT total ≥80) (Young 2016). Since the moderate users are often una-
ble to control their Internet use (Young 2016), we considered both moderate and exces-
sive use of Internet as problematic. is notion of PIU classification is fairly supported
by the existing literature (Al-Gamal etal. 2015; Ghamari etal. 2011; Kormas etal. 2011;
Lam and Peng 2010; Ni etal. 2009). Since the Bangla IAT score ranges from 18 to 90,
a score of 45 or above has been defined as PIU in the present study. e tool has been
tested in different studies with sound psychometric properties (Jelenchick etal. 2012;
Widyanto and McMurran 2004). e Bangla IAT and its factors have been found to
have sound reliability and strong convergent and discriminant validity (Karim and Nigar
2014). Internal consistency for this sample was found excellent (α=0.91).
Socio‑demographic andbehavioural measures
Demographic information consisting of age, sex, living arrangement (with family
vs dorm/mess), relationship status (single, partnered, separated) and perceived SES
were recorded through a separate demographic information recording sheet. SES was
Page 7 of 14
Islam and Hossin Asian J of Gambling Issues and Public Health (2016) 6:11
measured by asking the respondents to think about their society and respond where they
belong to in terms of education, money and job: upper class, middle class or lower class.
e health behavioural factors included in the study were physical activity, sleep dura-
tion and smoking. To assess physical activity, respondents were asked if they participate
in moderate physical activities for at least 10min at a time each day, such as brisk walk-
ing, cycling, swimming or any other activity that causes some increase in breathing or
heart rate. As for sleep duration, respondents’ were asked to report their average dura-
tion of sleep per day. Based on the US National Sleep Foundation’s recommendations,
sleep duration was categorized as normal (7–9h), short (<7 h) and long sleep (>9 h)
(Hirshkowitz etal. 2015). Participants also reported whether they were smoking ciga-
rette by answering ‘yes’ or ‘no’.
Statistical analyses
All data were analysed using the Statistical Package for Social Sciences (SPSS) version
20. Unadjusted prevalence percentages of PIU were obtained for the total sample and
for all variables of interest. Chi square test for independence was used to assess the dif-
ferences in the prevalence of PIU by socio-demographic and behavioural factors. e
associations between PIU, other explanatory variables and psychological distress were
examined at both bivariate and multivariate levels, using binary logistic regression. e
regression analyses were based on three models. e unadjusted odds ratios (OR) for
the associations between all explanatory variables and psychological distress were esti-
mated in model 1. Model 2 explored the association between PIU and psychological
distress, controlling for the socio-demographic variables. Model 3 represents the fully
adjusted model that further controlled for the behavioural factors. All point estimates
were reported with 95% confidence intervals.
Results
Table1 shows the distribution of the sample as well as the prevalence of PIU by socio-
demographic and behavioural factors. Nearly 24% of the total sample falls under the cat-
egory of PIU. e age of the study participants ranges from 20 to 30 with a mean of
25.1. Almost two-thirds of the participants were males (69.8%) out of which 28.8% were
classified as problematic Internet users. e proportion of PIU between male and female
differs significantly (p<0.001). Regarding SES, most of the participants reported to be
in the middle class (60.0%), a few were in low SES (15.4%) and the others belonged to
high SES (22.7%). e prevalence rates of PIU across SES categories indicate a negative
trend which means that the higher the SES, the lower the prevalence of PIU. is trend
is statistically significant at <0.05 level (p-for-trend: 0.018). Around 40% of the partic-
ipants were in partnered (romantic) relationship while 42.6% were single. A consider-
able proportion of the participants (17.1%) also reported to have split up their romantic
relationship. In general, the prevalence of PIU is more common in this group compared
to those who are single or partnered but the differences between them are statistically
significant only at <0.1 level. Around 70% of the study participants reported to be liv-
ing in either university or private dormitories while the remaining was living with their
families. e proportion of PIU was mostly similar across the living status of the partici-
pants. As regards the behavioural factors, PIU was higher among smokers compared to
Page 8 of 14
Islam and Hossin Asian J of Gambling Issues and Public Health (2016) 6:11
non-smokers (32.5 vs 21.8%). e difference of the proportion of the number of smok-
ers and non-smokers by PIU status was statistically significant (p<0.05). Compared to
those engaged in moderate or vigorous physical activity, the prevalence of PIU was also
significantly higher for those not engaged in physical activity (20.0 vs 27.9%). e dura-
tion of sleep of the study participants, however, did not show any statistical significance
in terms of the prevalence of PIU.
e prevalence of psychological distress among the study participants was 28.4% (data
not shown). e associations of psychological distress with PIU and other socio-demo-
graphic and behavioural correlates are presented in Table2. e crude ORs from the
logistic regression analyses demonstrate that PIU is associated with a 2.63-fold increased
odds of having psychological distress. When the socio-demographic factors were entered
in the model, the effect size became slightly smaller but still remained significant and
robust (OR 2.48, 95% CI 1.65, 3.72). Further statistical adjustment for the lifestyle fac-
tors did not attenuate the association considerably. As the multivariate analysis in model
3 suggests, PIU is stronglyassociated with a higher prevalence of psychological distress
(OR 2.37, 95% CI 1.57, 3.58) even when all other factors are held constant. Only a small
part (<10%) of the original association was accounted for by the socio-demographic and
behavioural factors considered in the analyses. However, of all the explanatory variables
included in the fully adjusted model, low SES emerged as the strongest predictor of psy-
chological distress, followed by PIU and smoking. e study results further show that
there is a clear socioeconomic gradient in psychological distress, with the individuals
at the bottom of the socioeconomic hierarchy reporting more psychological problems
than those at the top. For example, the individuals belonging to the low socioeconomic
class has 2.45 times elevated odds (95% CI 1.33, 4.50) of psychological distress relative to
the high socioeconomic class. e corresponding odds of psychological distress among
those identifying themselves as middle class is 1.62 (95% CI 1.01, 2.61). Furthermore,
the estimated odds of reporting psychological distress is significantly higher among the
smokers (OR 1.86, 95% CI 1.17, 2.96) compared to the non-smokers. No statistically sig-
nificant interactions, however, were found between PIU and other covariates in regard to
the prevalence of psychological distress.
Discussion
e study sought to report the prevalence of PIU among graduate students of Bang-
ladesh and its distribution by socio-economic and behavioural characteristics of the
participants. In addition, the study explored the association between PIU and psycholog-
ical distress controlling for the socio-demographic and behavioural covariates. Results
revealed that nearly 24% of the participants used Internet problematically. ere was a
significant sex difference in the prevalence of PIU with males reporting PIU more than
females. e prevalence rates of PIU also varied across socioeconomic status, relation-
ship status, smoking habit and physical activity. e study found that a strong and robust
association of PIU with psychological distress persists even when the socio-demographic
and behavioural factors are adjusted for.
e prevalence of PIU found in this study (23.9%) is lower than the rates obtained
in university samples of similar age groups from Muslim majority countries like Jor-
dan (40%) and Iran (39.6%) but higher than the British sample (18.3%) (Al-Gamal etal.
Page 9 of 14
Islam and Hossin Asian J of Gambling Issues and Public Health (2016) 6:11
2015; Ataee etal. 2014; Niemz etal. 2005). When both moderate and excessive users of
Internet are incorporated into PIU according to the IAT, the prevalence rates are usu-
ally higher. e prevalence figures from Asian countries affirms this assertion (Mak
etal. 2014). Furthermore, time could be another potential determinant of high preva-
lence. With time, the use of Internet increases, so does the PIU. Evidence from earlier
studies tended to indicate a lower prevalence than the most recent studies. Nonetheless,
the prevalence of PIU found for the Bangladeshi sample bears potential to draw clinical
attention as roughly one out of four students is problematic Internet user.
In agreement with previous literature, this study shows that males were more predis-
posed to PIU behaviour (Ha and Hwang 2014; Morahan-Martin and Schumacher 2000;
Table 2 Association betweenproblematic Internet use andpsychological distress, control-
ling forsocio-demographic andbehavioural correlates (n=573)
The ORs in italics indicate statistical signicance
Model 1: unadjusted
Model 2: adjusted for age, sex, socioeconomic status, relationship status, and living arrangement
Model 3: adjusted for model 2+smoking, physical activity, and sleep duration
OR odds ratio, CI condence interval
Explanatory variables Model 1 Model 2 Model 3
OR [95% CI] OR [95% CI] OR [95% CI]
Problematic Internet use
No [ref.] 1.00 1.00 1.00
Yes 2.63 [1.77, 3.90] 2.48 [1.65, 3.72] 2.37 [1.57, 3.58]
Age (years)
20–25 [ref.] 1.00 1.00 1.00
26–30 0.84 [0.59, 1.20] 0.79 [0.55, 1.15] 0.78 [0.53, 1.14]
Sex
Female [ref.] 1.00 1.00 1.00
Male 1.44 [0.98, 2.12] 1.14 [0.74, 1.74] 0.99 [0.63, 1.55]
Socioeconomic status
High [ref.] 1.00 1.00 1.00
Middle 1.60 [1.02, 2.53] 1.59 [1.00, 2.54] 1.62 [1.01, 2.61]
Low 2.55 [1.43, 4.55] 2.21 [1.22, 4.02] 2.45 [1.33, 4.50]
Relationship status
Single [ref.] 1.00 1.00 1.00
Partnered 1.02 [0.70, 1.50] 1.07 [0.72, 1.60] 1.02 [0.68, 1.54]
Separated 1.24 [0.76, 2.03] 1.16 [0.70, 1.93] 1.07 [0.64, 1.80]
Living arrangement
With family [ref.] 1.00 1.00 1.00
Dorm/mess 1.22 [0.83, 1.79] 1.22 [0.81, 1.84] 1.25 [0.83, 1.89]
Smoking
No [ref.] 1.00 1.00
Yes 1.87 [1.23, 2.84] 1.86 [1.17, 2.96]
Physical activity
No [ref.] 1.00 1.00
Yes .85 [0.58, 1.20] 0.84 [0.58, 1.21]
Sleep duration
Normal sleep [ref.] 1.00 1.00
Short sleep 0.81 [0.55, 1.20] 0.84 [0.56, 1.26]
Long sleep 1.89 [0.83, 4.33] 1.88 [0.79, 4.48]
Page 10 of 14
Islam and Hossin Asian J of Gambling Issues and Public Health (2016) 6:11
Shahnaz and Karim 2014). e sex difference in PIU is commonly explained by the dis-
tinct personality patterns of males and females and the purpose of using Internet. Males,
for instance, are usually more enthusiastic about exploring the unknown or discovering
new inventions (e.g., high on sensation seeking) (Ball etal. 1984). is, in turn, could
gradually lead to compulsive Internet use as Internet is currently the main hub of all
the exciting adventures. Secondly, males are found to engage more in addictive contents
such online gaming, pornography and cybersex compared to their female counterparts
(Bruno etal. 2014; Doornwaard etal. 2016; Young 1999). It was found that males score
low particularly in the time control factor of IAT, indicating a lack of control over time
when using Internet, a leading cause of PIU (Bruno etal. 2014).
SES is another strong risk factor of PIU. In this study, SES has been inversely asso-
ciated with PIU, that is, the lower the SES, the higher the PIU. is finding is also in
line with the evidence found in a Greek study which demonstrated that lower socio-
economic position was associated with higher score on the ‘neglect of social life’ fac-
tor of IAT (Andreou and Svoli 2013). Generally, the students from low socioeconomic
backgrounds migrate to the capital city from the rural parts of the country to study at
the premier public university, leaving behind their close relatives. ey are, therefore,
more likely to consider online engagement as a strategy to alleviate their social isolation
and loneliness. is online engagement, if not controlled, could appear as problematic
(Caplan 2007; Kim and Haridakis 2009). PIU was also found to be more common among
those who had a ‘breakup’ in their romantic relationship. Existing literature suggests that
split-up romantic relationship may cause poor mental health (Monroe etal. 1999; Simon
and Barrett 2010) by increasing frustration and social anxiety which, as a consequence,
might manifest as addictive behaviours like PIU (Liu and Kuo 2007). We also found that
PIU was higher among the participants who smoke cigarette and those not engaged in
physical activity. An earlier study focusing on an adolescent sample in China also found
non-involvement in physical activity as a risk factor for PIU, although smoking was not
significantly associated with IA (Lam etal. 2009). Based on a nationally representative
sample, Lee etal. (2013), however, found that smoking predisposes to a high risk of IA
(OR 1.203, p=0.004) even after adjusting for sex, age, stress, depressed mood and sui-
cidal ideation. It is possible that smoking and IA share similar bio-psycho-social mecha-
nisms—a potential area for further investigation.
e study results further demonstrate that PIU is significantly associated with psycho-
logical distress, a finding which is largely in agreement with previous studies (Al-Gamal
etal. 2015; Cotten etal. 2011; Ha and Hwang 2014; Shapira etal. 2000; Tsai etal. 2009).
We, however, anticipated that this association would be substantially attenuated once the
socio-demographic and behavioural factors are adjusted for, given the well-established
links between these factors and psychological wellbeing (Bayram and Bilgel 2008; Gloz-
ier etal. 2010; Matsuzaki etal. 2007). To our surprise, only an insignificant part of the
observed association was explained by these factors, despite the fact that SES and smok-
ing appeared as strong predictors of psychological distress in the fully adjusted regres-
sion model. ere could be several possible explanations for that. One crucial feature
of PIU is the lack of control over the use of Internet. us, one may end up in distress
after spending increasing amount of time online. e individual may promise not does it
again but later get engaged in the same addictive behaviour, leading to more frustration
Page 11 of 14
Islam and Hossin Asian J of Gambling Issues and Public Health (2016) 6:11
and distress. It thus goes like a vicious cycle. e psychological distress, therefore, might
arise due to the negative consequences this cycle brings on the academic performance
(Jiang 2014), unhealthy lifestyle such as irregular meal and bedtime (Nigar and Karim
2014) and interpersonal relationships (Cao etal. 2011). Moreover, in the case of ado-
lescents, a common interpretation of PIU is that the problematic users use Internet as
a way of escaping from the real life stress or challenges (Kim and Davis 2009). ere is
a high possibility that graduate students also over engage themselves in online activities
in order to cope with the strain associated with their career choice—a common stressor
during this period. is is the time when the graduates struggle to shift to working life
from student life. Since the unemployment rate is high in Bangladesh, many of them
remain uncertain about their future career position. Responding to real life situation in
this way, i.e., by over engaging online, however, could manifest as poor psychological
health such as distress. Because PIU is kind of maladaptive coping strategy that limits
or cuts down on actual physical activity or involvement, both are crucial for psycho-
logical well-being. Furthermore, PIU co-morbids with other mental health conditions
such as ADHD, social anxiety, phobia, depression, loneliness and low self-esteem (Cao
etal. 2011; Liberatore etal. 2011). e simultaneous presence of two or more health
conditions can influence one another in terms of symptoms moderation. As mentioned
earlier, psychological distress in this study has been conceptualized as changes in affec-
tive and somatic symptoms relative to usual levels of health and hence PIU might affect
other possible co-morbid conditions which potentially result in psychological distress
within the individual. A closer look at these issues could unearth the underlying possible
mechanism of this association.
e findings of our study, however, have to be cautiously interpreted as it has some
limitations. Firstly, due to the nature of cross-sectional design, we cannot infer a causal
connection between PIU and psychological distress. at is, the link between PIU and
psychological distress as observed in this study is simply associational and hence we
could not rule out the possibility that psychological distress causes PIU rather than the
other way round. A longitudinal study design would be more appropriate to unravel the
possible causal direction of the association. Secondly, the study sample was derived from
one university only. Although, the sampled university is the largest populated academia
in the country, incorporating participants from other universities could more accurately
reflect the scenario under investigation. irdly, as indicated above, we did not consider
some potential confounding factors such as the participants’ academic pressure, the
pressure to find a job, the purpose of using Internet, loneliness and comorbid mental
health problems which are likely to bias the association of interest.
Despite these limitations, the study findings carry important implications for planning
public health policies and interventions. Our study findings, for example, imply that the
males, low socioeconomic groups, smokers and the physically inactive young population
can be the important target groups for any intervention designed toprevent problem-
atic use of Internet. Besides, the findings of this study can aid the mental health service
providers in their clinical practices, especially when it comes to dealing with the patients
with IA behaviour. Regardless of the causal direction of the association between PIU and
psychological distress, the mental health professionals such as university counsellors or
psychotherapists are expected to pay attention to the signs and symptoms of PIU during
Page 12 of 14
Islam and Hossin Asian J of Gambling Issues and Public Health (2016) 6:11
the assessment and formulation of psychological health issues of the patients as PIU is
strongly associated with their psychological wellbeing. However, further research is war-
ranted for a broader and more in-depth understanding of the link between PIU and psy-
chological distress taking into account the comorbid mental health conditions, academic
performance, psycho-social stressors and interpersonal problems of the young adults of
Bangladesh.
Authors’ contributions
Both MAI and MZH designed the study. MAI conducted the field survey. MZH analyzed the data and wrote the results
section. MAI drafted the manuscript. Both authors read and approved the final manuscript.
Author details
1 Department of Educational and Counselling Psychology, University of Dhaka, Dhaka 1000, Bangladesh. 2 Department
of Public Health Sciences, Karolinska Institute, Stockholm, Sweden.
Competing interests
Both authors declare that they have no competing interests.
Received: 28 July 2016 Accepted: 21 November 2016
References
Al-Gamal, E., Alzayyat, A., & Ahmad, M. M. (2015). Prevalence of internet addiction and its association with psychologi-
cal distress and coping strategies among university students in Jordan. Perspectives in Psychiatric Care, 52, 49–61.
doi:10.1111/ppc.12102.
Andreou, E., & Svoli, H. (2013). The association between internet user characteristics and dimensions of internet addic-
tion among Greek adolescents. International Journal of Mental Health and Addiction, 11(2), 139–148. doi:10.1007/
s11469-012-9404-3.
APA. (2013). Diagnostic and statistical manual of mental disorders: DSM-5 (5th ed.). Washington DC: American Psychiatric
Association.
Ataee, M., Ahmadi Jouybari, T., Emdadi, S., Hatamzadeh, N., Mahboubi, M., & Aghaei, A. (2014). Prevalence of Internet
addiction and its associated factors in Hamadan University of medical college students. Life Science Journal, 11(4),
214–217.
Ball, I. L., Farnill, D., & Wangeman, J. F. (1984). Sex and age differences in sensation seeking: Some national comparisons.
British Journal of Psychology, 75(2), 257–265. doi:10.1017/CBO9781107415324.004.
Bayram, N., & Bilgel, N. (2008). The prevalence and socio-demographic correlations of depression, anxiety and stress
among a group of university students. Social Psychiatry and Psychiatric Epidemiology, 43(8), 667–672. doi:10.1007/
s00127-008-0345-x.
Bruno, A., Scimeca, G., Cava, L., Pandolfo, G., Zoccali, R. A., & Muscatello, M. R. A. (2014). Prevalence of internet addiction
in a sample of Southern Italian high school students. International Journal of Mental Health and Addiction, 12(6),
708–715. doi:10.1007/s11469-014-9497-y.
BTRC. (2016). Internet subscribers in Bangladesh April, 2016. Dhaka: Government of Bangladesh. http://www.btrc.gov.bd/
content/internet-subscribers-bangladesh-april-2016.
Cao, H., Sun, Y., Wan, Y., Hao, J., & Tao, F. (2011). Problematic Internet use in Chinese adolescents and its relation to psycho-
somatic symptoms and life satisfaction. BMC Public Health, 11, 802. http://ovidsp.ovid.com/ovidweb.cgi?T=JS&PAGE
=reference&D=emed11&NEWS=N&AN=21995654.
Caplan, S. E. (2007). Relations among loneliness, social anxiety, and problematic internet use. CyberPsychology & Behavior,
10(2), 234–242. doi:10.1089/cpb.2006.9963.
Cheng, C., & Li, A. Y.-L. (2014). Internet addiction prevalence and quality of (real) life: A meta-analysis of 31 nations across
seven world regions. Cyberpsychology, Behavior and Social Networking, 17(12), 755–760. doi:10.1089/cyber.2014.0317.
Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure of perceived stress. Journal of Health and Social Behav-
ior, 24(4), 385–396. http://www.ncbi.nlm.nih.gov/pubmed/6668417.
Cotten, S. R., Goldner, M., Hale, T. M., & Drentea, P. (2011). The importance of type, amount, and timing of internet use for
understanding psychological distress. Social Science Quarterly, 92(1), 119–139. doi:10.1111/j.1540-6237.2011.00760.x.
Doornwaard, S. M., van den Eijnden, R. J. J. M., Baams, L., Vanwesenbeeck, I., & ter Bogt, T. F. M. (2016). Lower psychological
well-being and excessive sexual interest predict symptoms of compulsive use of sexually explicit internet material
among adolescent boys. Journal of Youth and Adolescence, 45(1), 73–84. doi:10.1007/s10964-015-0326-9.
Ghamari, F., Mohammadbeigi, A., Mohammadsalehi, N., & Hashiani, A. A. (2011). Internet addiction and modeling its risk
factors in medical students, Iran. Indian Journal of Psychological Medicine, 33(2), 158. doi:10.4103/0253-7176.92068.
Glozier, N., Martiniuk, A., Patton, G., Ivers, R., Li, Q., Hickie, I., et al. (2010). Short sleep duration in prevalent and persistent
psychological distress in young adults. Sleep, 33(9), 1139–1145.
Goldberg, D. P. (1972). The detection of psychiatric illness by questionnaire. London: Oxford University Press.
Goldberg, D. P., Oldehinkel, T., & Ormel, J. (1998). Why GHQ threshold varies from one place to another. Psychological
Medicine, 28(4), 915–921. doi:10.1017/S0033291798006874.
Goodchild, M. E., & Duncan-Jones, P. (1985). Chronicity and the General Health Questionnaire. The British Journal of Psy-
chiatry, 146(1), 55–61. http://bjp.rcpsych.org/content/146/1/55.abstract.
Page 13 of 14
Islam and Hossin Asian J of Gambling Issues and Public Health (2016) 6:11
Ha, Y.-M., & Hwang, W. J. (2014). Gender differences in internet addiction associated with psychological health indicators
among adolescents using a national web-based survey. International Journal of Mental Health and Addiction, 12(5),
660–669. doi:10.1007/s11469-014-9500-7.
Hirshkowitz, M., Whiton, K., Albert, S. M., Alessi, C., Bruni, O., DonCarlos, L., et al. (2015). National sleep foundation’s sleep
time duration recommendations: Methodology and results summary. Sleep Health, 1(1), 40–43. doi:10.1016/j.
sleh.2014.12.010.
Jelenchick, L. A., Becker, T., & Moreno, M. A. (2012). Assessing the psychometric properties of the internet addiction test
(IAT) in US college students. Psychiatry Research, 196(2–3), 296–301. doi:10.1016/j.psychres.2011.09.007.
Jiang, Q. (2014). Internet addiction among young people in China: Internet connectedness, online gaming, and aca-
demic performance decrement. Internet Research, 24(1), 2–20. doi:10.1108/IntR-01-2013-0004.
Karim, A. K. M. R., & Nigar, N. (2014). The internet addiction test: Assessing its psychometric properties in Bangladeshi
culture. Asian Journal of Psychiatry, 10, 75–83. doi:10.1016/j.ajp.2013.10.011.
Kim, H. K., & Davis, K. E. (2009). Toward a comprehensive theory of problematic Internet use: Evaluating the role of
self-esteem, anxiety, flow, and the self-rated importance of Internet activities. Computers in Human Behavior, 25(2),
490–500. doi:10.1016/j.chb.2008.11.001.
Kim, J., & Haridakis, P. M. (2009). The role of Internet user characteristics and motives in explaining three
dimensions of Internet addiction. Journal of Computer-Mediated Communication, 14(4), 988–1015.
doi:10.1111/j.1083-6101.2009.01478.x.
Ko, C. H., Yen, J. Y., Yen, C. F., Chen, C. S., & Chen, C. C. (2012). The association between Internet addiction and psychiatric
disorder: A review of the literature. European Psychiatry, 27(1), 1–8. doi:10.1016/j.eurpsy.2010.04.011.
Kormas, G., Critselis, E., Janikian, M., Kafetzis, D., & Tsitsika, A. (2011). Risk factors and psychosocial characteristics of poten-
tial problematic and problematic internet use among adolescents: A cross-sectional study. BMC Public Health, 11(1),
595. doi:10.1186/1471-2458-11-595.
Kuss, D. J., Griffiths, M. D., Karila, L., & Billieux, J. (2013a). Internet addiction: A systematic review of epidemiological
research for the last decade. Current Pharmaceutical Design, 1(4), 397–413. doi:10.2174/13816128113199990617.
Kuss, D., Rooij, A. Van, & Shorter, G. (2013b). Internet addiction in adolescents: Prevalence and risk factors. Computers in
Human Behavior, 29(5), 1987–1996. doi:10.1016/j.chb.2013.04.002.
Lam, L. T., & Peng, Z.-W. (2010). Effect of pathological use of the internet on adolescent mental health: A prospective
study. Archives of Pediatrics and Adolescent Medicine, 164(10), 901–906. doi:10.1001/archpediatrics.2010.159.
Lam, L. T., Peng, Z., Mai, J., & Jing, J. (2009). Factors associated with Internet addiction among adolescents. CYBERPSYCHOL-
OGY & Behavior, 12(5), 551–555. doi:10.1089/cpb.2009.0036.
Lanthier, R. P., & Windham, R. C. (2004). Internet use and college adjustment: The moderating role of gender. Computers in
Human Behavior, 20(5), 591–606. doi:10.1016/j.chb.2003.11.003.
Lee, Y. S., Han, D. H., Kim, S. M., & Renshaw, P. F. (2013). Substance abuse precedes internet addiction. Addictive Behaviors,
38(4), 2022–2025. doi:10.1530/ERC-14-0411.Persistent.
Liberatore, K. A., Rosario, K., Colón-De Martí, L. N., & Martínez, K. G. (2011). Prevalence of Internet addiction in Latino
adolescents with psychiatric diagnosis. Cyberpsychology, Behavior and Social Networking, 14(6), 399–402. doi:10.1089/
cyber.2010.0252.
Liu, C.-Y., & Kuo, F.-Y. (2007). A study of Internet addiction through the lens of the interpersonal theory. CYBERPSYCHOLOGY
& Behavior, 10(6), 799–804. doi:10.1089/cpb.2007.9951.
Mak, K.-K., Lai, C.-M., Watanabe, H., Kim, D.-I., Bahar, N., Ramos, M., et al. (2014). Epidemiology of internet behaviors and
addiction among adolescents in six Asian countries. Cyberpsychology, Behavior and Social Networking, 17(11),
720–728. doi:10.1089/cyber.2014.0139.
Marmot, M. (2009). Social determinants of health inequalities. Lancet, 365(9464), 1099–1104. doi:10.1016/
S0140-6736(05)71146-6.
Matsuzaki, I., Sagara, T., Ohshita, Y., Nagase, H., Ogino, K., Eboshida, A., et al. (2007). Psychological factors including sense of
coherence and some lifestyles are related to general health questionnaire-12 (GHQ-12) in elderly workers in Japan.
Environmental Health and Preventive Medicine, 12(2), 71–77. doi:10.1007/BF02898152.
Monroe, S. M., Rohde, P., Seeley, J. R., & Lewinsohn, P. M. (1999). Life events and depression in adolescence: Relationship
loss as a prospective risk factor for first onset of major depressive disorder. Journal of Abnormal Psychology, 108(4),
606–614. doi:10.1037/0021-843X.108.4.606.
Morahan-Martin, J., & Schumacher, P. (2000). Incidence and correlates of pathological internet use among college stu-
dents. Computers in Human Behavior, 16(1), 13–29. doi:10.1016/S0747-5632(99)00049-7.
Morgan, C., & Cotten, S. R. (2003). The relationship between internet activities and depressive symptoms in a sample of
college freshmen. CyberPsychology & Behavior, 6(2), 133–142. doi:10.1089/109493103321640329.
Ni, X., Yan, H., Chen, S., & Liu, Z. (2009). Factors influencing internet addiction in a sample of freshmen university students
in China. Cyberpsychology & Behavior, 12(3), 327–330. doi:10.1089/cpb.2008.0321.
Niemz, K., Griffiths, M., & Banyard, P. (2005). Prevalence of pathological Internet use among university students and cor-
relations with self-esteem, the General Health Questionnaire (GHQ), and disinhibition. CyberPsychology & Behavior,
8(6), 562–570. doi:10.1089/cpb.2005.8.562.
Nigar, N., & Karim, A. K. M. R. (2014). Impact of internet addiction on social intimacy and lifestyle of young adults. Dhaka
Univeristy Journal of Psychology, 38, 99–110.
Ostovar, S., Allahyar, N., Aminpoor, H., Moafian, F., Nor, M. B. M., & Griffiths, M. D. (2016). Internet addiction and its psycho-
social risks (depression, anxiety, stress and loneliness) among Iranian adolescents and young adults: A structural
equation model in a cross-sectional study. International Journal of Mental Health and Addiction, 14(3), 257–267.
doi:10.1007/s11469-015-9628-0.
Perrin, A. (2015). Social networking usage: 2005-2015. Pew Research Center. http://www.pewinternet.org/2015/10/08/
social-networking-usage-2005-2015/
Poli, R., & Agrimi, E. (2012). Internet addiction disorder: Prevalence in an Italian student population. Nordic Journal of
Psychiatry, 66(1), 55–59. doi:10.3109/08039488.2011.605169.
Page 14 of 14
Islam and Hossin Asian J of Gambling Issues and Public Health (2016) 6:11
Rhoades, G. K., Dush, C. M. K., Atkins, D. C., Stanley, S. M., & Markman, H. J. (2011). Breaking up is hard to do: The impact
of unmarried relationship dissolution on mental health and life satisfaction. Journal of Family Psychology, 25(3),
366–374. doi:10.1016/j.surg.2006.10.010.Use.
Sabatini, F., & Sarracino, F. (2014). Online networks and subjective well-being (MPRA Paper No. 56436). University Library of
Munich.
Shahnaz, I., & Karim, A. K. M. R. (2014). Gender difference in internet addiction among young adults. Dhaka Univeristy
Journal of Psychology, 38, 111–122.
Shapira, N. A., Goldsmith, T. D., Keck, P. E., Khosla, U. M., & McElroy, S. L. (2000). Psychiatric features of individuals with prob-
lematic internet use. Journal of Affective Disorders, 57(1–3), 267–272. doi:10.1016/S0165-0327(99)00107-X.
Simon, R. W., & Barrett, A. E. (2010). Nonmarital romantic relationships and mental health in early adulthood:
Does the association differ for women and men? Journal of Health and Social Behavior, 51(2), 168–182.
doi:10.1177/0022146510372343.
Sorcar, N. R., & Rahman, A. (1990). Occupational stress, marital status and job satisfaction of working women. The Dhaka
University Studies, XI(1), 55–61.
Tsai, H. F., Cheng, S. H., Yeh, T. L., Shih, C. C., Chen, K. C., Yang, Y. C., et al. (2009). The risk factors of internet addiction—A
survey of university freshmen. Psychiatry Research, 167(3), 294–299. doi:10.1016/j.psychres.2008.01.015.
Wallace, P. (2014). Internet addiction disorder and youth. EMBO Reports, 15(1), 12–16. doi:10.1002/embr.201338222.
Widyanto, L., & McMurran, M. (2004). The psychometric properties of the internet addiction test. CyberPsychology & Behav-
ior, 7(4), 443–450. doi:10.1089/cpb.2004.7.443.
Young, K. S. (1996). Internet addiction: The emergence of a new clinical disorder. CyberPsychology & Behavior, 1(3),
237–244. doi:10.1089/cpb.1998.1.237.
Young, K. S. (1998). Caught in the net: How to recognize the signs of internet addiction—And a winning strategy for recovery.
New York: Wiley.
Young, K. S. (1999). Internet addiction: Symptoms, evaluation, and treatment. Innovations in Clinical Practice, 17, 19–31.
doi:10.1007/s10879-009-9120-x.
Young, K. S. (2004). Internet addiction: A new clinical phenomenon and its consequences. American Behavioral Scientist,
48(4), 402–415. doi:10.1177/0002764204270278.
Young, K. S. (2016). Internet addiction test manual. Bradford, PA: Center for Internet Addiction Recovery. http://netaddic-
tion.com/.
Yu, L., & Shek, D. T. L. (2013). Internet addiction in Hong Kong adolescents: A three-year longitudinal study. Journal of
Pediatric and Adolescent Gynecology, 26(3 Suppl.), S10–S17. doi:10.1016/j.jpag.2013.03.010.