(Open Access) The Association between Pathological Internet Use and Comorbid Psychopathology: A Systematic Review. (2013) | Vladimir Carli

Q: What have the authors contributed in "The association between pathological internet use and comorbid psychopathology: a systematic review" ?

The primary objective of this systematic review was to identify and evaluate studies performed on the correlation between PIU and comorbid psychopathology ; the secondary aims were to map the geographical distribution of studies, present a current synthesis of the evidence, and assess the quality of available research. An electronic literature search was conducted using the following databases: MEDLINE, PsycARTICLES, PsychINFO, Global Health, and Web of Science. Com/psp V. C. and T. D. contributed equally to this article and therefore both should be considered as first authors.

Fax +41 61 306 12 34

E-Mail karger@karger.ch

www.karger.com

Review

Psychopathology 2013;46:1–13

DOI: 10.1159/000337971

The Association between Pathological

Internet Use and Comorbid

Psychopathology: A Systematic Review

V. Carli

T. Durkee

D. Wasserman

G. Hadlaczky

R. Despalins

E. Kramarz

C. Wasserman

b, d

M. Sarchiapone

C.W. Hoven

R. Brunner

M. Kaess

e, f

National Centre for Suicide Research and Prevention of Mental Ill-Health (NASP) at Karolinska Institutet,

Stockholm , Sweden;

Department of Child and Adolescent Psychiatry, Columbia University-New York State

Psychiatric Institute, and

Department of Epidemiology, Mailman School of Public Health, Columbia University,

New York, N.Y. , USA;

Department of Health Sciences, University of Molise, Campobasso , Italy;

Section for

Disorders of Personality Development, Department of Child and Adolescent Psychiatry, Center of Psychosocial

Medicine, University of Heidelberg, Heidelberg , Germany;

Orygen Youth Health, Melbourne, Vic. , Australia

and psychopathology, including depression, anxiety, symp-

toms of attention deficit and hyperactivity disorder (ADHD),

obsessive-compulsive symptoms, social phobia and hostili-

ty/aggression. Effect sizes for the correlations observed were

identified from either the respective publication or calculat-

ed using Cohen’s d or R

. The potential effect of publication

bias was assessed using a funnel plot model and evaluated

by Egger’s test based on a linear regression. Results: The ma-

jority of research was conducted in Asia and comprised

cross-sectional designs. Only one prospective study was

identified. Twenty articles met the preset inclusion and ex-

clusion criteria; 75% reported significant correlations of PIU

with depression, 57% with anxiety, 100% with symptoms of

ADHD, 60% with obsessive-compulsive symptoms, and 66%

with hostility/aggression. No study reported associations

between PIU and social phobia. The majority of studies re-

ported a higher rate of PIU among males than females. The

relative risks ranged from an OR of 1.02 to an OR of 11.66. The

strongest correlations were observed between PIU and de-

Key Words

Pathological Internet use ⴢ Internet addiction ⴢ

Psychopathology ⴢ Depression ⴢ Anxiety ⴢ Attention

deficit and hyperactivity disorder ⴢ Obsessive-compulsive

symptoms ⴢ Social phobia ⴢ Hostility/aggression

Abstract

Background: Pathological Internet use (PIU) has been con-

ceptualized as an impulse-control disorder that shares char-

acteristics with behavioral addiction. Research has indicated

a potential link between PIU and psychopathology; how-

er, the significance of the correlation remains ambiguous.

The primary objective of this systematic review was to iden-

tify and evaluate studies performed on the correlation be-

tween PIU and comorbid psychopathology; the secondary

aims were to map the geographical distribution of studies,

present a current synthesis of the evidence, and assess the

quality of available research. Sampling and Methods: An

electronic literature search was conducted using the follow-

ing databases: MEDLINE, PsycARTICLES, PsychINFO, Global

Health, and Web of Science. PIU and known synonyms were

included in the search. Data were extracted based on PIU

Received: July 26, 2011

Accepted after revision: February 25, 2012

Published online: July 31, 2012

Michael Kaess, MD

Orygen Youth Health

35 Poplar Road

Parkville, VIC 3052 (Australia)

Tel. +61 3 9342 2800, E-Mail michael.kaess

 @  unimelb.edu.au

0254–4962/13/0461–0001$38.00/0

Accessible online at:

www.karger.com/psp

V.C. and T.D. contributed equally to this article and therefore both

should be considered as first authors.

Downloaded by:

Umea University Library

130.239.20.133 - 2/22/2018 10:17:19 PM

Carli et al.

Psychopathology 2013;46:1–13

pression; the weakest was hostility/aggression. Conclusions:

Depression and symptoms of ADHD appeared to have the

most significant and consistent correlation with PIU. Asso-

ciations were reported to be higher among males in all age

groups. Limitations included heterogeneity in the definition

and diagnosis of PIU. More studies with prospective designs

in Western countries are critically needed.

Introduction

Internet use has grown substantially over the past few

decades, accounting for nearly 2 billion users globally

[1] .

Although studies indicate that the majority of Internet

users are among adolescents and young adults

[2–4] , re-

search shows that Internet usage is rising among older

populations

[5] . Given the broad spectrum of Internet

users, it is important to understand the potential risks

involved in compulsive use. Public health concerns are

emerging concerning the propensity of compulsive Inter-

net use developing into pathological behaviors.

The pathway from adaptive to pathological Internet

use (PIU) appears to be ambiguous

[6] ; however, there

are noteworthy characteristics distinguishing the two

groups. Among adaptive users, the Internet appears to

serve as a supportive tool, rather than a source of identity.

There is evidence showing that adaptive use facilitates

new and existing relationships through frequent and ac-

cessible online communication; it can promote socializa-

tion and self-esteem

[7] , as well as decrease loneliness [8] .

Conversely, evidence has indicated that pathological In-

ternet users tend to spend more hours online compared to

adaptive users (e.g. online 6 10–20 h/week)

[9, 10] and are

prone to use the Internet for specific online activities (e.g.

compulsive gambling

[11] , pornography [12–14] , extreme

role-playing fantasies

[15 –17] , and excessive gaming [18] ).

Research suggests that PIU may not only reflect a risk-

behavioral syndrome, but also a clinical disorder, due to

the presence of withdrawal and tolerance symptoms

[19] .

PIU is conceptually modeled as an impulse-control

disorder that does not involve an intoxicant, and it shares

qualities related to behavioral addiction

[20] . Behavioral

addiction, as suggested by Griffiths

[21] , is a paradigm

that is based on the Diagnostic and Statistical Manual

of Mental Disorders (DSM-IV) criteria for pathological

gambling and substance-dependence; it comprises six

explicit traits: salience, mood modification, tolerance,

withdrawal symptoms, conflict and relapse. It is hypoth-

esized that the respective attributes of behavioral addic-

tion are present among pathological Internet users

[22–

24]

. On this basis, PIU has been proposed for inclusion in

the DSM-V as a behavioral addiction, but without success

[25, 26] . Instead, PIU will be inserted in the DSM-V ap-

pendix, stipulating that more research is required before

a diagnosis can be incorporated into the DSM nosological

system

[27] ; this could potentially be a contributing factor

to the lack of a universal diagnostic criteria for PIU, in-

evitably influencing outcomes of PIU-related studies.

The prevalence of PIU varies among populations. Ep-

idemiological studies have reported considerable varia-

tions in the prevalence of PIU among adolescents and

young adults, ranging from 0.9 to 37.9%

[28–31] in Asia

and 2

[27] to 18.3% [32–36] in Europe. In the US general

population, PIU prevalence varies between 0.3

[37] and

8.1%

[38] . It is evident that there are extreme variances in

PIU prevalence across countries and cultures; further

analysis on the psychological effect of prevalent PIU is

necessary.

There may be severe mental and emotional implica-

tions for those with PIU; it should be noted that PIU may

also occur as a consequence of ongoing mental health is-

sues. Research has indicated a potential correlation be-

tween PIU and impulsivity

[39, 40] , depression [41, 42] ,

anxiety

[43] , psychosis [44] , obsessive-compulsive symp-

toms

[45] , and social anxiety/phobia [46] ; however, data

so far has been contradictory with regards to comorbid

psychopathology. To the best of our knowledge, an evi-

dence-based systematic review examining the associa-

tion between PIU and psychopathological traits is still

lacking

[47] ; scientific-based outcomes are required for

preventive and treatment efforts.

The primary aim of this systematic review was to

identify and evaluate studies performed on the correla-

tion between PIU and comorbid psychopathology; sec-

ondary aims were to assess the diagnostic criteria for

measuring PIU and outcome measures of psychopathol-

ogy, map the geographical distribution of studies, and

evaluate levels of evidence. Based on the available litera-

ture, the following psychopathologies were included: de-

pression

[48] , anxiety [49] , symptoms of attention deficit

and hyperactivity disorder (ADHD)

[50] , obsessive-

compulsive symptoms

[51] , social phobia [52] , and hos-

tility/aggression

[53] .

M e t h o d s

An electronic literature search was conducted using the fol-

lowing databases: MEDLINE, PsycARTICLES, PsychINFO,

Global Health, and Web of Science. There were no restrictions on

Downloaded by:

Umea University Library

130.239.20.133 - 2/22/2018 10:17:19 PM

Association between PIU and Comorbid

Psychopathology

Psychopathology 2013;46:1–13

language, time, or publication status. Key identifiers used were

‘Internet addiction’ or ‘Internet addiction disorder’ or ‘Internet

dependency’ or ‘pathological Internet use’ or ‘problematic Inter-

net use’ or ‘compulsive Internet use’ or ‘excessive Internet use’ or

‘computer addiction’, combined with the identifiers ‘depression’

or ‘anxiety’ or ‘obsessive-compulsive’ or ‘ADHD’ or ‘social phobia’

or ‘hostility’ or ‘aggression’.

Articles were systematically and independently reviewed by

the authors; assessments were performed regarding the study

type, study population, methodology, outcome measures, effect

sizes, and interpretation of results. The inclusion criteria for stud-

ies involved population-based studies with a large sample size

( 1 200 subjects), ascertained diagnostic criteria for PIU, subse-

quent reporting on the correlation between PIU and predeter-

mined psychopathologies, and the psychometric outcome mea-

sures assessing psychopathology. Studies were excluded if there

were no clear diagnostic criteria, a significant sampling bias or

small sample size ( ! 200 subjects), only focused on specific sub-

types of PIU (e.g. compulsive online gambling), and/or were case

studies and/or treatment assessment.

Studies were rated according to the scheme proposed by the

Oxford Centre for Evidence-Based Medicine Results

[54] and

evaluated by the following criteria: observation of a full or par-

tial association, significance level, and adjustments for con-

founders. Full association was considered when a correlation

was found for both sexes after multivariate analyses. If a correla-

tion was identified for only one gender, it was classified as a par-

tial association. The geographical distribution of studies was

also mapped.

Effect size of the associations was identified by either the orig-

inal publications or calculated using the data of the respective

publications. Identified effect sizes were reported mostly in odds

ratios (OR), with one publication reporting in hazard ratios (HR);

the calculated effect sizes were either Cohen’s d or R

. In order to

compare the different associations, the effect sizes d and R

were

stated as small, moderate, or large, according to Cohen

[55] ; OR

were converted into these groups according to Chinn

[56] . The

effect sizes were interpreted accordingly: small (d = 0.2, R

= 0.01,

OR = 1.45), moderate (d = 0.5, R

= 0.06, OR = 2.50), and large

(d = 0.8, R

= 0.14, OR = 4.25).

The potential effect of publication bias was assessed for the

relationship between PIU and depression. This effect could not

be estimated for other psychopathologies, given that too few pub-

lications met the inclusion criteria. The publication bias was ex-

plored for depression by a funnel plot model. This graph was cre-

ated by plotting the log OR against the standard error of these

measures

[57] . A funnel plot graph, shaped with a symmetrical

distribution, would indicate no publication bias, whereas an

asymmetric plot would suggest bias; this could be due to unre-

ported studies, small sample sizes, or low significance levels

[58] .

The asymmetry of the funnel plot was statistically evaluated by

Egger’s test, which is based on a linear regression of the normal-

ized effect estimate (estimate divided by its standard error)

against precision (reciprocal of the standard error)

[59] . If a pub-

lication bias was found, a trim and fill method

[60] was used to

estimate the number of missing studies and adjusted according-

ly. This met hod is commonly used to remove t he asymmetric side

of the funnel plot by artificially imputing the missing studies,

based on the other side of the graph.

R e s u l t s

After deleting duplicate studies, a total of 185 articles

were screened and identified through the present system-

atic search. As a result, 32 studies were immediately ex-

cluded, as they were based on other Internet-related top-

ics. Twenty articles were included in the respective sys-

tematic review in accordance with the inclusion and

exclusion criteria ( fig.1 )

[61] .

Description of Included Studies

The included articles comprised one prospective co-

hort study (level of evidence: 1B)

[62] , two case-control

studies (level of evidence: 3B)

[63, 64] , and 17 cross-sec-

tional studies (level of evidence: 4)

[65–81] , as illustrated

in table1 . Half of the studies (n = 10) in this review tar-

geted adolescent groups, seven studies targeted young

adults, and three studies were aimed at the general popu-

lation; all studies examined both genders.

Geographical Distribution of Studies

Overall, the majority of studies performed on PIU

were implemented in Asian countries. Eleven studies

were performed in China

[62, 64–68, 70, 73, 76, 77, 81]

and five in South Korea

[72, 74, 75, 79, 82] . The remaining

studies were conducted in the US

[71] , UK [63] , Norway

[69] , and Turkey [80] .

Methods Assessing PIU

There are no standardized diagnostic criteria for iden-

tifying PIU; however, there are several assessment instru-

ments that are often utilized in PIU research. The most

common psychometric instrument(s) for measuring PIU

is the Young’s Diagnostic Questionnaire for Internet Ad-

diction (YDQ)

[83] . The YDQ is built upon the DSM-IV

diagnostic criteria for pathological gambling and has

been employed and validated in other studies

[32, 35] . In

the YDQ, the diagnosis is based on a pattern of Internet

usage that results in a clinical impairment or distress in

accordance to the presence of the following criteria: (1)

preoccupation with the Internet; (2) need for longer

amounts of time online to achieve satisfaction; (3) repeat-

ed unsuccessful efforts to control, cut back, or stop Inter-

net use; (4) restlessness, moodiness, depression, or irrita-

bility when attempting to cut down or stop Internet use;

(5) staying online longer than originally intended; (6)

jeopardizing or risking the loss of a significant relation-

ship, job, or educational opportunity because of the In-

ternet; (7) lying to family members, therapists, or others

to conceal the extent of involvement with the Internet;

Downloaded by:

Umea University Library

130.239.20.133 - 2/22/2018 10:17:19 PM

Carli et al.

Psychopathology 2013;46:1–13

and (8) using the Internet as a way of escaping from prob-

lems or relieving a dysphoric mood

[20, 83] . The respec-

tive eight criteria are evaluated through eight ‘yes’ or ‘no’

questions with a total score ranging from 0–8. Those

scoring 6 5 were classified as pathological. In the present

review, two studies used the YDQ to measure PIU

[64,

69]

. Based on the YDQ, Young [84] further modified the

assessment instrument to measure severity, thereby es-

tablishing a 20-item questionnaire that measures mild,

moderate, and severe levels of PIU. Referred to as the In-

ternet Addiction Test (IAT), the psychometric properties

of this instrument have been evaluated and ascertained

as valid and reliable

[85] . The IAT was the most utilized

assessment of PIU taxonomy in this review, with eight

studies

[63, 65, 68, 72, 74, 75, 78, 79] reporting to have

used it. The second most frequently utilized instrument

was the Chen Internet Addiction Scale

[86] , which was

reported in seven studies

[62, 66, 67, 70, 73, 77, 81] in this

analysis. The Chen Internet Addiction Scale is a 26-item

questionnaire, which has also been validated, and assess-

es five dimensions of the condition: compulsive use,

withdrawal, tolerance, problems with interpersonal rela-

tionships, and time management

[87] . The three residual

studies employed atypical measures: Internet Usage

Questionnaire

[71] , Diagnostic Criteria of Internet Ad-

diction

[76] , and Problematic Internet Use Scale [80] .

Methods Assessing Psychopathology

Measurements of psychopathology in the scrutinized

studies were performed by different psychometric mech-

anisms. To measure depression, five studies

[62, 66, 71–

73] used the Center for Epidemiologic Studies Depression

Scale; two studies

[63, 80] used Beck’s Depression Inven-

tory; and two studies

[65, 68] used the Zung Self-Rating

Depression Scale. One study employed the Diagnostic In-

terview Schedule for Children with Major Depression

Disorder

[74] ; one study utilized the Mini-International

Neuropsychological Interview

[76] ; one study used a

modified Diagnostic Scale of Excessive Internet Use,

which included an assessment of depression

[79] ; and one

study used the 12-item version of the General Health

Questionnaire

[80] . To measure ADHD symptoms, five

studies

[62, 73, 76–78] used diverse versions of the Adult

ADHD Self-Report Scale. To measure anxiety, one study

used the Self-Rating Anxiety Scale

[68] , and one study

used single-item questions

[69] for both anxiety and ob-

Records identified through

database searching

(n = 195)

Records after duplicates removed

(n = 185)

Records excluded (based on

The Association between Pathological Internet Use and Comorbid Psychopathology: A Systematic Review.

Figures

Citations

The relationship between addictive use of social media and video games and symptoms of psychiatric disorders: a large-scale cross-sectional study

LETTERS to the EDITORPublication bias

The association between internet addiction and psychiatric co-morbidity: a meta-analysis

Working towards an international consensus on criteria for assessing Internet Gaming Disorder: a critical commentary on Petry et al (2014)

Epidemiology of internet behaviors and addiction among adolescents in six Asian countries.

References

Statistical Power Analysis for the Behavioral Sciences

Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement

Bias in meta-analysis detected by a simple, graphical test

Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement

Trim and fill: A simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis.

Related Papers (5)

Internet Addiction: The Emergence of a New Clinical Disorder

The Comorbid Psychiatric Symptoms of Internet Addiction: Attention Deficit and Hyperactivity Disorder (ADHD), Depression, Social Phobia, and Hostility

Internet addiction: a systematic review of epidemiological research for the last decade.

Issues for DSM-V: internet addiction.

A cognitive-behavioral model of pathological Internet use

Frequently Asked Questions (1)

Q1. What have the authors contributed in "The association between pathological internet use and comorbid psychopathology: a systematic review" ?