Clinical Psychology Review 92 (2022) 102128

Contents lists available at ScienceDirect

Clinical Psychology Review

journal homepage: www.elsevier.com/locate/clinpsychrev

Review

Global prevalence of digital addiction in general population: A systematic

review and meta-analysis
Shi-Qiu Meng a, 1, Jia-Lu Cheng a, 1, Yang-Yang Li a, 1, Xiao-Qin Yang a, Jun-Wei Zheng a,
Xiang-Wen Chang a, Yu Shi b, Yun Chen b, Lin Lu a, c, Yan Sun a, *, Yan-Ping Bao a, *, Jie Shi a, b, d, e, *
a
National Institute on Drug Dependence and Beijing Key Laboratory of Drug Dependence, Peking University, Beijing, China
b
Peking University Shenzhen Hospital, Shenzhen, China
c
Peking University Sixth Hospital, Peking University Institute of Mental Health, NHC Key Laboratory of Mental Health (Peking University), National Clinical Research
Center for Mental Disorders (Peking University Sixth Hospital), Beijing, China
d
The State Key Laboratory of Natural and Biomimetic Drugs, Peking University, Beijing, China
e
The Key Laboratory for Neuroscience of the Ministry of Education and Health, Peking University, Beijing, China

A R T I C L E I N F O A B S T R A C T

Keywords: The present meta-analytic review aimed to synthesize the global prevalence characteristics of digital addiction in
Digital addiction the general population. We searched PubMed, Embase, Cochrane Library, and PsycINFO for studies reporting
Behavioral addiction prevalence of various subtypes of digital addiction published before October 31, 2021. Studies were eligible if
Internet addiction
they were published in peer-reviewed journals, used a validated tool to assess digital addiction, and passed the
Gaming addiction
qualify assessment. In total, 498 articles with 507 studies were included in systematic review, and the meta-
Prevalence
Meta-analysis analysis included 495 articles with 504 studies covering 2,123,762 individuals from 64 countries. Global
pooled prevalence estimates were 26.99% (95% CI, 22.73–31.73) for smartphone addiction, 17.42% (95% CI,
12.42–23.89) for social media addiction, 14.22% (95% CI, 12.90–15.65) for Internet addiction, 8.23% (95% CI,
5.75–11.66) for cybersex addiction, and 6.04% (95% CI, 4.80–7.57) for game addiction. Higher prevalence of
digital addiction was found in Eastern Mediterranean region and low/lower-middle income countries. Males had
higher risk for Internet and game addiction. An increasing trend of digital addiction during the past two decades
was found, which dramatically worsened during COVID-19 pandemic. This study provides the first and
comprehensive estimation for the global prevalence of multiple subtypes of digital addiction, which varied be-
tween regions, economic levels, time periods of publication, genders, and assessment scales.
PROSPERO ID: CRD42020171117.

1. Introduction Notably, digital addiction does not necessarily involve Internet use, and
thus it includes not only addiction to online activities, but also addiction
The number of active digital users was 4.66 billion for Internet, 4.32 to offline activities using digital devices, such as offline game addiction
billion for mobile Internet, and 4.2 billion for social media by January (Almourad, McAlaney, Skinner, Pleya, & Ali, 2020; Christakis, 2019).
2021 (Statista, 2021), with an average online time of 6.7 h daily. Previous evidence showed that digital addiction caused significant im-
However, persistent or recurrent use of digital media, referring to digital pairments in health, study, work, and other social functions, and marked
devices (e.g., computers, smartphone) and related activities (e.g., distress in personal, family, and social well-being (Bell, Bishop, &
games, social media), could lead to digital addiction (Christakis, 2019; Przybylski, 2015; Dahl & Bergmark, 2020; WHO, 2014). Concerns about
WHO, 2014). Digital addiction is an umbrella term incorporating sub- increased risks for digital addiction during the COVID-19 pandemic have
types of the long-standing problem of Internet addiction, the highly- also been raised (Király et al., 2020). To provide basis for formulating
discussed issue of game addiction, and the emerging topic on social prevention and treatment measures, it is urgent for systematically esti-
media addiction, or other digital media addiction (Christakis, 2019). mating the prevalence characteristics and contributors for digital

* Corresponding authors at: National Institute on Drug Dependence, Peking University, 38 Xueyuan Road, Haidian District, Beijing 100191, China.
E-mail addresses: sunyan@bjmu.edu.cn (Y. Sun), baoyp@bjmu.edu.cn (Y.-P. Bao), shijie@bjmu.edu.cn (J. Shi).
1
Equally contributed to this work

https://doi.org/10.1016/j.cpr.2022.102128
Received 26 September 2021; Received in revised form 29 December 2021; Accepted 20 January 2022
Available online 25 January 2022
0272-7358/© 2022 Elsevier Ltd. All rights reserved.
S.-Q. Meng et al. Clinical Psychology Review 92 (2022) 102128

addiction (Potenza, Higuchi, & Brand, 2018). reference lists of identified studies and relevant reviews. We excluded
Epidemiological studies showed the prevalence of different subtypes studies based on selective sample, reviews, conference abstracts and
of digital addiction ranging from 0.5% to 84%, with great variations in theses. For multiple articles using data from the same population, the
subtypes of digital media, diagnostic tools, and methodological quality, one with the largest sample size and the most comprehensive results was
and bias from small sample size, single population, non-random sam- included. We contacted authors to request lacking or additional data.
pling, and invalidated screening methods (Alhassan et al., 2018; Feng, Four researchers independently screened on each included study. Dis-
Ramo, Chan, & Bourgeois, 2017). Previous systematic reviews had only crepancies were adjudicated through discussion with the review team.
summarized the prevalence of a single subtype of digital addiction (e.g.,
Internet addiction (Cheng & Li, 2014; Kuss, Griffiths, Karila, & Billieux,
2.2. Data extraction and quality assessment
2014; Pan, Chiu, & Lin, 2020; Su, Han, Jin, Yan, & Potenza, 2019;
Weinstein & Lejoyeux, 2010), game addiction (Ferguson, Coulson, &
For each study, data were independently extracted using a stan-
Barnett, 2011; Mihara & Higuchi, 2017; Pan et al., 2020; Stevens,
dardized scale, and cross-checked by four researchers. The non-English
Dorstyn, Delfabbro, & King, 2020), or smartphone addiction (Sahu,
records were translated. The prevalence estimates in this study were
Gandhi, & Sharma, 2019; Sohn, Rees, Wildridge, Kalk, & Carter, 2019)),
calculated by dividing the number of people screened as “addiction” by
and focused on specific populations in specific regions (e.g., Chinese
the number of valid respondents.
college students (Li et al., 2018); Shao et al., 2018), American youths
The quality of each study was assessed by a modified version of
(Moreno, Jelenchick, Cox, Young, & Christakis, 2011), Southeast Asian
Newcastle-Ottawa Scale (NOS) (Stang, 2010) from five dimensions:
(Chia et al., 2020), Iranian (Modara et al., 2017), people from Gulf
sample representativeness, sample size, response rate, ascertainment of
countries (Al-Khani et al., 2021), healthcare professionals (Buneviciene
digital addiction, and thoroughness of descriptive data reporting (Ap-
& Bunevicius, 2020), medical students (Zhang, Lim, Lee, & Ho, 2018),
pendix D in the Supplement). Each dimension has a score of 0 or 1, and a
and adolescents (Fam, 2018; Paulus, Ohmann, von Gontard, & Popow,
total score of 0 to 2 was considered as having a high risk in quality. Four
2018)). Until now, it still lacks a comprehensive estimate for the prev-
researchers independently scored on each included study, and then
alence characteristics of different subtypes of digital addiction in the
discrepancies were discussed with the review team. The studies with
general population worldwide, which could provide critical information
high risk in quality (NOS score < 3) were excluded.
for addressing the global status of digital addiction and benefit the
formulation of coping strategies, especially under the impact of COVID-
19. 2.3. Statistical analysis
Thus, we conducted a systematic review and meta-analysis to make
comprehensive estimate of the prevalence of 5 subtypes of digital The pooled prevalence and 95% confidence interval (95% CI) of
addiction in the general population at a global level, and display the digital addiction were estimated. Multiple types of digital addiction
differences between sub-populations. We also aimed to investigate the were involved in this analysis, including Internet addiction, game
effects of demographic, geographic, and clinical characteristics, as well addiction, smartphone addiction, social media addiction, and cybersex
as other associated factors on the prevalence of multiple subtypes of addiction. The variance of the raw prevalence from each included study
digital addiction. was stabilized using the logit transformation before pooling the preva-
lence estimates. All estimates were presented after back transformation
2. Methods (Lipsey & Wilson, 2000). Heterogeneity of the pooled prevalence among
studies was assessed using Cochrane Q and I2 index. For the Cochrane Q
2.1. Search strategy and selection criteria test, P < 0.05 represented significant heterogeneity. For I2 index, values
of 25% or less correspond to low degrees of heterogeneity, 26% to 50%
This systematic review and meta-analysis was conducted according to moderate degrees, and greater than 50% to high degrees (Higgins &
to the Preferred Reporting Items for Systematic Reviews and Meta- Thompson, 2002; Higgins, Thompson, Deeks, & Altman, 2003). As ex-
Analysis (PRISMA, Appendix A-B in the Supplement) (Moher, Liberati, pected, high heterogeneity was observed, so a random-effects model
Tetzlaff, & Altman, 2009). The protocol of the study was preregistered (DerSimonian and Laird method) meta-analysis was used to calculate
(https://www.crd.york.ac.uk/PROSPERO/, PROSPERO ID: the pooled prevalence and 95% CIs (Deeks, Higgins, & Altman, 2019). A
CRD42020171117). leave-one-out sensitivity analysis was applied for each meta-analysis.
We searched PubMed, Embase, Cochrane Library, and PsycINFO to Funnel plots and the Egger’s test were performed to assess publication
identify relevant literature on the prevalence of digital addiction in the bias when more than 10 studies were available in a single meta-analysis
general population published before October 31, 2021. We used search (Egger, Davey Smith, Schneider, & Minder, 1997; Sterne & Egger, 2001).
items related to digital devices or activities (internet OR digital OR We performed subgroup meta-analyses for Internet addiction, game
screen OR cyber* OR net OR online OR media OR electronic device* OR addiction, smartphone addiction, and social media addiction. Stratifi-
electronic gadgets OR computer OR mobile OR phone OR smartphone cation analysis was layered for WHO region (WHO, 2021b), World Bank
OR television OR TV OR video OR facebook OR game OR gaming), region (World Bank, 2021), time period of publication (each five year as
problematic use (addict* OR use OR dependen* OR overuse OR abuse a period), gender, age group (<12 years child/12–18 years adolescent/
OR disorder OR excessive OR effects OR habits OR misuse OR patho- ≥18 years adult), education level (elementary school/secondary school/
logical OR problem* OR compulsive OR heavy), and prevalence (prev- university), way of investigation (online/off-line), sample size (<1000/
alence OR survey OR rate OR scale OR screening OR situation OR ≥1000), the score of each criterion in NOS (0/1), and screening tool. As
epidemic OR epidemiological OR occurrence OR investigation). Full for study selection, only commonly used screening tools or commonly
searching details for each database and subtypes of digital addiction used cut-off values were synthesized (“commonly used” refers to those
included were listed (Appendix C in the Supplement). screening tools used in at least 5 studies).
The inclusion criteria were as follows: studies published on peer- Meta-regression analyses were conducted to explore the source of
reviewed journals; cross-sectional studies or longitudinal studies heterogeneity of the pooled prevalence estimates (Jackson, Law, Rücker,
providing baseline data; investigations based on general population; & Schwarzer, 2017). The selected characteristics included WHO region,
definitions and measurement tools of general or subtypes of digital World Bank region, time period of publication, gender, age group, ed-
addiction were clearly reported; directly providing the prevalence rate ucation level, way of investigation, screening tool and score of the
or sufficient data to calculate the prevalence. No language restriction overall NOS. Data analysis was done using R (version 4.0.0) with the
was applied. Additional records were identified through screening of “dmetar” package.

2
S.-Q. Meng et al. Clinical Psychology Review 92 (2022) 102128

3. Results response rate of over 95%. 414 (82.14%) had qualified descriptive sta-
tistics with proper measures of dispersion. 494 (98.02%) fulfilled the
3.1. Study selection and characteristics ascertainment criterion by using a validated tool to screen digital
addiction (eTable 14 in the Supplement).
37,965 records were identified through our searching strategy
(Fig. 1). After applying the eligibility criteria, 498 eligible articles with a
total of 507 independent studies were included in this systematic re- 3.2. Internet addiction
view. Across 507 studies, 341 studies provided prevalence data on
Internet addiction, 78 on game addiction, 82 on smartphone addiction, Pooled prevalence of Internet addiction from 341 studies (n =
31 on social media addiction, 6 on cybersex addiction, 2 on online 1,631,543) was 14.22% (95% CI, 12.90–15.65; Fig. 2). Geographically,
gambling, and 1 on general digital addiction (Fig. 2-5, eTables 1–6 in the the prevalence of Internet addiction was the highest in the WHO African
Supplement). The sample sizes of the included studies ranged from 119 region (34.53%; 95% CI, 30.46–38.83), followed by the Eastern Medi-
to 223,542 with a total number of 2,127,059. The studies covered 64 terranean region (30.11%; 95% CI, 23.44–37.74). By contrast, the esti-
countries in 6 WHO regions: 7 (1.30%) studies from 4 countries in the mates were relatively lower in the region of the Americas (11.06%; 95%
African region, 33 (6.12%) studies from 6 countries in the region of the CI, 8.0–15.19) and European region (11.06%; 95% CI, 9.31–13.09, P <
Americas, 52 (9.65%) studies from 11 countries in the Eastern Medi- 0.001; Fig. 6A). In terms of economic status, World Bank regions at low
terranean region, 177 (32.83%) studies from 26 countries in the Euro- income level had the highest prevalence estimates (25.15%; 95% CI,
pean region, 44 (8.16%) studies from 7 countries in the South-East Asia 15.60–37.93; P < 0.001). From 1999 to 2021, the prevalence of Internet
region, and 224 (41.56%) studies from 10 countries in the Western addiction significantly increased with the year of publication (slope =
Pacific region (eTable 7 in the Supplement). Since we analyzed each 7.54% per year increase; 95% CI,1.25–60.34; P < 0.001, eTable 15 in the
subtype of digital addiction separately, the 2 studies of online gambling Supplement). When stratified by investigation time, pooled prevalence
and the 1 study of general digital addiction were not included in the of 11 studies that were related to COVID-19 increased significantly
meta-analysis. As a result, 495 eligible articles (504 studies) were compared with previous studies (32.39% vs 15.20%, p < 0.0001; eTa-
included in the quantitative meta-analysis (n = 2,123,762). ble 16 in the Supplement). The prevalence of Internet addiction among
All the 504 included studies had a total quality score of at least 3 males (17.15%; 95% CI, 14.89–19.68) was higher than females (11.60%;
(eTables 8–13 in the Supplement). 193 (38.29%) studies met the crite- 95% CI, 9.89–13.57; P = 0.0003). Sample sizes of less than 1000 par-
rion for sample representativeness (i.e., surveying in multiple in- ticipants had higher prevalence (17.71%; 95% CI, 15.11–20.66) than
stitutions, or multiple groups of people). 495 (98.21%) fulfilled the those of 1000 or more participants (12.23%; 95% CI, 10.87–13.74; p =
sample size criterion (n > 300). 116 (23.02%) provided satisfactory 0.0002, Fig. 2). Significant difference was observed among various
comparability between respondents and non-respondents, or had a screening instruments (p < 0.001, Fig. 2). The 20-item Young’s Internet
Addiction Test (YIAT-20) was the most widely used screening scale for

Fig. 1. Flowchart.

3
S.-Q. Meng et al. Clinical Psychology Review 92 (2022) 102128

Fig. 2. Prevalence of Internet addiction using random-effects meta-analysis and subgroup meta-analysis.
Note: CIAS-26 = the 26-item Chen’s Internet Addiction Scale; CERI-10 = the 10-item ‘Questionnaire for Mobile Phone-related Experiences’ (Cuestionario de
Experiencias Relacionadas con el Movil); CIUS-14 = the 14-item Compulsive Internet Use Scale; YDQ-8 = the 8-item Young’s Diagnostic Questionnaire; YIAT-20 =
the 20-item Young’s Internet Addiction Test; NA = not applicable.

Internet addiction, and 40/70 or 50/80 were commonly used cut-off of Internet addiction when stratified by other characteristics (Fig. 2).
values to identify moderate/severe addiction. Specifically, the preva-
lence of moderate and severe Internet addiction assessed by the 40/70 3.3. Game addiction
cut-off value of YIAT-20 was 34.03% and 3.38%, while 50/80 cut-off
revealing lower prevalence estimates of 17.93% and 1.43% with a The pooled prevalence of game addiction from 78 studies (n =
more stringent threshold (eTable 17 in the Supplement). Subgroup 261,100) was 6.04% (95% CI, 4.80–7.57; Fig. 3). Region of the Americas
analysis indicated no statistically significant difference in the prevalence had the highest prevalence of game addiction (9.85%; 95% CI,

4
S.-Q. Meng et al. Clinical Psychology Review 92 (2022) 102128

Fig. 3. Prevalence of game addiction using random-effects meta-analysis and subgroup meta-analysis.
Note: DSM-5 = scales derived from the fifth version of Diagnostic and Statistical Manual of Mental Disorders criteria for Internet gaming disorder; DSM-IV = scales
derived from the fourth version of Diagnostic and Statistical Manual of Mental Disorders criteria for gambling disorder or substance use disorder; NA =
not applicable.

4.90–18.82), followed by Western Pacific region (9.26%; 95% CI, criteria for Internet Gaming Disorder, and scales derived from the DSM-
7.15–11.92), whereas the European region had the lowest prevalence IV criteria for gambling disorder or substance use disorder. The DSM-IV-
(4.27%; 95% CI, 2.90–6.25, P = 0.0115; Fig. 6B). The prevalence peaked adapted scales were the most commonly applied tools in studies before
during 2005–2009 (11.76%; 95% CI 11.07–12.48), declined during 2015 (8 in 14 studies, 57.14%), yet were largely replaced by DSM-5-
2010–2014 (5.92%; 95% CI, 3.55–9.71), then remained stable over time based scales after 2015 (44 in 64 studies, 69.75%, data not shown).
(5.10% during 2015–2019, 95% CI 3.67–7.04), and increased after 2020 Studies using scales based on DSM-5 and DSM-IV criteria did not display
(8.33%, 95% CI 5.52–12.41). Males (10.71%; 95% CI, 8.22–13.85) had significant differences in prevalence (5.02% vs 8.00%, p = 0.1095). The
significantly higher estimates than females (4.19%; 95% CI, 2.96–5.90; prevalence of game addiction did not differ significantly when stratified
p < 0.0001). Commonly used screening instruments of game addiction by other characteristics (Fig. 3).
could be classified into two categories - scales adapted from the Diag-
nostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5)

5
S.-Q. Meng et al. Clinical Psychology Review 92 (2022) 102128

Fig. 4. Prevalence of smartphone addiction using random-effects meta-analysis and subgroup meta-analysis.
Note: CERI-10 = the 10-item ‘Questionnaire for Mobile Phone-related Experiences’ (Cuestionario de Experiencias Relacionadas con el Movil); SAS-SV = the 10-item
Smartphone Addiction Scale-Short Version; S-Scale-15 = the Korean Scale for Smartphone Addiction for Adolescents and Adults; NA = not applicable.

3.4. Smartphone addiction 23.02–62.99) and the Eastern Mediterranean region (38.88%; 95% CI,
26.14–53.34), and relatively low in the European region (18.51%; 95%
Regarding smartphone addiction, the pooled prevalence from 82 CI, 11.44–28.54; Fig. 6C). World Bank regions at lower-middle income
studies (n = 147,319) was 26.99% (95% CI, 22.73–31.73; Fig. 4). level had the highest prevalence estimates (43.84%; 95% CI,
Similar to Internet addiction, the prevalence of smartphone addiction 30.65–57.95; p = 0.0003). The reported prevalence estimates of
was relatively high in the South-East Asia region (41.63%; 95% CI, smartphone addiction were mainly published after 2015 (74 studies,

6
S.-Q. Meng et al. Clinical Psychology Review 92 (2022) 102128

Fig. 5. Prevalence of social media addiction using random-effects meta-analysis and subgroup meta-analysis.
Note: BSMAS-6 = the 6-item Bergen Social Media Addiction Scale; BFAS-6 = the 6-item Bergen Facebook Addiction Scale; NA = not applicable.

90.24%), and studies published during 2020 to 2021 had higher rate and children (15.19%; 95% CI, 13.86–16.61; p < 0.0001). Among the
(34.5%; 95% CI, 27.52–42.22) than those published during 2015 to three screening instruments used in more than 5 studies, significant
2019 (25.96%; 95% CI, 20.42–32.39), 2010 to 2014 (9.99%; 95% CI, differences in prevalence estimates were observed (P < 0.0001). Sub-
4.82–19.57), and 2005 to 2009 (15.04%; 95% CI, 11.75–19.04). Uni- group analysis indicated no statistically significant difference in the
versities have relatively high prevalence rate (26.96%; 95% CI, prevalence of smartphone addiction by other characteristics (Fig. 4).
20.84–34.11) compared to secondary schools (19.61%; 95% CI,
13.82–27.08) and elementary schools (15.19%; 95% CI, 13.86–16.61; p
= 0.0002). Prevalence among adults (26.84%; 95% CI, 20.92–33.73) 3.5. Social media addiction
was higher compared with adolescents (21.62%; 95% CI, 15.88–28.74)
The pooled prevalence of social media addiction from 31 studies (n

7
S.-Q. Meng et al. Clinical Psychology Review 92 (2022) 102128

A. Internet addiction B. Game addiction

C. Smartphone addiction D. Social media addiction

Fig. 6. Geographic distribution of different types of digital addiction (A. Internet addiction; B. Game addiction; C. Smartphone addiction; D. Social media addiction).

= 72,561) was 17.42% (95% CI, 12.42–23.89; Fig. 5). Geographically, score = 0) showed higher prevalence of game addiction than those with
similar to smartphone addiction, the prevalence in the South-East Asia more representative sample (NOS sample representativeness score = 1)
region was relatively high (59.36%; 95% CI, 30.37–83.03), and the (7.93% vs 4.32%; P = 0.0070; Fig. 3). Sensitivity analysis of each type of
prevalence in the region of the Americas was relatively low (11.66%; digital addiction showed that no single study had an excessive influence
95% CI, 5.72–22.33; P < 0.001; Fig. 6D). Also, the prevalence was on the pooled prevalence (eFigures 2–5 in the Supplement).
higher in lower-middle income regions (63.83%; 95% CI, 37.02–84.12)
compared with high income (16.12%; 95% CI, 11.52–22.10) and upper- 4. Discussion
middle income regions (21.26%; 95% CI, 15.15–29.00; P = 0.0009).
Most of the related studies were published after 2015 (28 studies, To our knowledge, the present study is the first systematic review
90.32%). The prevalence based on the 6-item Bergen Facebook Addic- and meta-analysis to comprehensively estimate the worldwide pooled
tion Scale (BFAS-6, 42.46%; 95% CI, 27.18–59.33) was significantly prevalence of multiple subtypes of digital addiction in general popula-
higher (p = 0.0046) than that based on the 6-item Bergen Social Media tion, based on available data from inception of each database to October
Addiction Scale (BSMAS-6, 14.85%; 95% CI, 7.78–26.51). No significant 31, 2021. Our findings showed the pooled estimate of prevalence across
difference was observed in the prevalence of social media addiction by subtypes of digital devices or related activities with substantial vari-
other characteristics (Fig. 5). ability. More importantly, we provided meta-analytic evidence that
COVID-19 exacerbated the increasing trend of Internet addiction. For
3.6. Cybersex addiction Internet, smartphone, and social media addiction, the prevalence varied
with different assessments. Higher prevalence of Internet addiction, and
The prevalence of cybersex addiction was reported in only 6 studies, smartphone addiction were found in Eastern Mediterranean region,
so subgroup analysis was not conducted. The estimates ranged from 4% while higher prevalence of game addiction was found in region of the
to 15%, with a pooled prevalence of 8.23% (95% CI, 5.75–11.66). All Americas. Participants from countries with low/lower-middle income
studies on cybersex addiction were from Europe. levels had higher burden of digital addiction behavior. Males tended to
have higher prevalence of Internet addiction and game addiction than
females. Other possible confounding factors included sample size, study
3.7. Publication bias and sensitivity analysis
quality, and ways of investigation on the prevalence. The present study
provided basic information for the prevention of digital addiction
No publication bias was observed according to visual inspection of
globally. Early and comprehensive prevention and intervention strate-
funnel plot and Egger’s test for Internet addiction (p = 0.18), smart-
gies for digital addiction are urgently needed worldwide, especially
phone addiction (p = 0.57), and social media addiction (p = 0.89,
under the influence of COVID-19, which has aggregated the prevalence
eFigure 1 in the Supplement). As for game addiction, bias assessment
situation of digital addiction. Research priorities lie in development of
showed asymmetry in the funnel plot (Egger’s test P < 0.001, eFigure 1
appropriate screening tools, reveal of neural mechanisms, epidemio-
in the Supplement). This bias was possibly related to the facts that
logical studies in less-developed regions, and investigation of effective
studies with less representative sample (NOS sample representativeness

8
S.-Q. Meng et al. Clinical Psychology Review 92 (2022) 102128

therapies. prevalence of Internet addiction than previous meta-analyses may be

that we included latest studies which displayed increasing prevalence,
4.1. Geographical distribution of digital addiction as well as non-English studies, from four major databases.
In the past decades, rapid penetration of Internet and development of
Most subtypes of digital addiction (Internet, game, smartphone, and technologies worldwide allowed easier access to computers, smart-
social media) had relatively high prevalence estimates in the Eastern phones, iPads, and various online applications (e.g., video games,
Mediterranean region, while had relatively low prevalence in the Eu- Facebook, and Twitter). Our results showed that the prevalence of
ropean region and the Americas (except for game addiction). Similarly, Internet addiction maintained a long-term growth during the past two
previous meta-analyses showed high prevalence rates of Internet decades. Consistently, a meta-analysis of 113 epidemiologic studies
addiction in the Middle East region, and low prevalence in Northern and covering 693,306 subjects published from 1996 to 2018 also found that
Western Europe, and revealed inverse association between prevalence of prevalence of Internet addiction increased over time (Pan et al., 2020).
Internet addiction and low quality of real life (Cheng & Li, 2014; Modara This rings an alarm for the policymakers to formulate effective strategies
et al., 2017). Participants with economic or social disadvantages are to reduce Internet addiction. Smartphone and social media addiction
more likely to use digital media to build connections, self-medicate, emerged and grew during the past decade. The prevalence of smart-
reduce stress, and alleviate mood as an escape from the unsatisfied phone addiction published during 2020 to 2021 was the highest. The
outer environment (Jang & Ji, 2012; Lopez-Fernandez, 2018), which is problematic smartphone use, especially among university students
consistent with our findings that the prevalence of digital addiction was needs more attention. However, consistent with findings in other meta-
relatively high in the World Bank low or lower-middle income regions. analysis (Feng et al., 2017; Pan et al., 2020), in contrast to the expanding
However, regional distribution in game addiction was different from market of online and off-line videogames, we did not find an increasing
other subtypes. Higher prevalence of game addiction was found in re- trend in game addiction during the past decades, which may result from
gion of the Americas and Western Pacific region, which mainly included the imbalanced sample sizes and methodological differences between
the United States, Canada, Brazil, Mexico, China, South Korea, and subgroups (Fam, 2018). Specifically, the three included studies (Grusser,
Singapore in the present study. The estimated prevalence of game Thalemann, Albrecht, & Thalemann, 2005; Grusser, Thalemann, &
addiction in our study was higher than that reported in a meta-analysis Griffiths, 2007; Li, Wang, & Wang, 2008) in 2005–2009 subgroup with
with studies from 2009 to 2019 (Stevens et al., 2020), which included relatively small sample sizes did not use screening scales adapted from
more European studies and fewer American and Asian studies. This the DSM-IV criteria as most studies during 2010–2014 did, which might
regional distribution characteristics in our study were in line with some lead to an overestimated prevalence for 2005–2009. The prevalence of
previous meta-analyses (Fam, 2018; Long et al., 2018; Stevens et al., game addiction from 2010 to 2014 to 2015–2019 was relatively stable,
2020) and had been linked to rapid development of game-related in- similar with the trend reported in a review based on national repre-
dustry, well-acceptance of game culture, and prevailing of e-sports in sentative samples from 2007 to 2016 (Feng et al., 2017). After the issue
these American and Asian countries (Chung, Sum, Chan, Lai, & Cheng, of DSM-5 in 2013, measures of game addiction were mainly based on
2019; Humphreys, 2019). Besides cultural factors, the regional dispar- criteria of Internet gaming disorder in DSM-5, which is stricter and tends
ities of prevalence of digital addiction may also be attributed to differ- to come to a lower prevalence estimate compared with previous mea-
ences in sample representativeness. For example, the prevalence of sures according to our results. Therefore, this relatively stable trend was
Internet addiction in the WHO African region were found to be higher possibly due to the combined effects of the booming game market and
than the Eastern Mediterranean region, while they were only based on the changes in screening and diagnostic methods.
small samples of community adults (Tiego et al., 2019) and university Starting from December 2019, the COVID-19 pandemic has caused
students (Asrese & Muche, 2020; Mboya et al., 2020). Similarly, social serious threat to the whole world. Home quarantine during the
media addiction exhibited the highest prevalence in the South-East Asia pandemic has given people more time and opportunities to access the
region, yet only a sample of Thai high school students (Khumsri, Yin- digital devices for work, study, and entertainment. Furthermore, the
gyeun, Manwong, Hanprathet, & Phanasathit, 2015) and two samples of pandemic and quarantine induced widespread psychological problems,
Bangladeshi university students were included (Jahan et al., 2021; which increased risks for development of addictive behaviors (Li et al.,
Mamun & Griffiths, 2019). Samples in these studies are not represen- 2021; Sun et al., 2020; Wang et al., 2021). Studies since 2020 have re-
tative of the whole population, and thus the findings of these regions ported that the prevalence of digital addiction have dramatically grown
should be taken with caution. This also revealed a gap to be filled that under the impact of the COVID-19 pandemic (Dong, Yang, Lu, & Hao,
estimates were lack in low and lower-middle income countries in the 2020; Islam et al., 2020; Li et al., 2021; Lin, 2020; Siste et al., 2020; Sun
African region and South-East Asia region. Compared with the high- et al., 2020), consistent with findings of our study. The present meta-
income countries, these regions are likely to have higher prevalence of analysis included 11 surveys on Internet addiction related to COVID-
digital addiction and fewer support on public mental health (WHO, 19, and found that the prevalence rates dramatically increased
2021a), thus highlighting a need to prioritize relevant studies in these compared with non-pandemic-related surveys (eTable 16 in the Sup-
regions. plement). These findings emphasized the importance and necessity of
prevention and intervention of digital addiction during the pandemic.
4.2. Prevalence trend of digital addiction over time Although some guidelines have been released to reduce use of digital
devices and prevent Internet addiction (Meng et al., 2020; WHO, 2020),
In the present study, we included studies before October 31, 2021, population-specific strategies for different subtypes of digital addiction
and found that global pooled prevalence was 26.99% for smartphone need to be developed during and post the pandemic.
addiction, 14.22% for Internet addiction, and 6.04% for game addiction.
In comparison with previous studies, a meta-analysis on Internet 4.3. Differences between screening tools
addiction including research before 2018 obtained a prevalence of
7.02% (Pan et al., 2020), and another with research before 2012 got a Shifting definition of subtypes of digital addiction had driven het-
prevalence of 6.0% (Cheng & Li, 2014), suggesting an increasing trend erogeneity in prevalence estimates. Regarding Internet addiction, the
of Internet addiction over time; prevalence estimates of game addiction included studies applied over 40 screening scales and showed a sur-
vary between 3.05% to 6.0% (Ferguson et al., 2011; Stevens et al., prising amount of variability. The most commonly used tool was YIAT-
2020); meta-analysis of smartphone addiction had only been conducted 20, which was adapted from DSM-IV criteria for gambling disorder and
for children and young people with a median prevalence of 23.3% (Sohn alcohol use disorder. Yet, as a screening scale instead of diagnostic
et al., 2019). The reason why the present study obtained higher pooled criteria, YIAT-20 had been reported to have limited clinical utility for

9
S.-Q. Meng et al. Clinical Psychology Review 92 (2022) 102128

evaluating the severity of Internet addiction (Ha, Yu, Park, Ryu, & Kim, Role of funding sources
2012). As for game addiction, DSM-5-based screening tools are more
strict and suitable to clinical diagnosis compared with other screening This work was supported by funding from the Ministry of Science and
tools, and resulted in lower prevalence estimates (Fam, 2018). We also Technology of China (2021ZD0202100, 2021ZD0201900,
found a similar trend that studies using DSM-5-based screening tools 2021YFF0306500), National Natural Science Foundation of China
tended to show lower prevalence than those with DSM-IV-based tools, (U1802283, 82130040, 81901352, 82171488), Beijing Municipal Sci-
but the differences were not statistically significant. For smartphone ence & Technology Commission (Z181100001518005), and Special
addiction, the 15-item Korean Smartphone Addiction Scale (S-Scale-15) Research Fund of PKUHSC for Prevention and Control of COVID-19
was only used in studies in South Korea, the 10-item Questionnaire for (BMU2020HKYZX008).
Mobile Phone-related Experiences (CERM-10) was only applied in
Europe and the Americas, and the Smartphone Addiction Scale-Short Contributors
Version (SAS-SV) was applied worldwide. For social media addiction,
BSMAS-6 and BFAS-6 were used for general and specific social media SQM, JLC, YYL, YS, YPB, and JS designed the study. SQM, YS, YPB,
addiction, respectively, and we found that the prevalence based on the LL, and JS oversaw its implementation. SQM, JLC, YYL, XQY, and JWZ
BFAS-6 was significantly higher than that based on the BSMAS-6. Thus, conducted literature search. SQM, JLC, YYL, JWZ, XQY, XWC, YS, and
the discrepancies between these scales possibly resulted from regional YC performed study selection, data extraction, and quality assessment.
disparity or their applicable subtypes. In addition, only one study used a JLC, SQM, YYL, and YPB designed and conducted data analysis. SQM,
scale for general digital addiction (Hawi, Samaha, & Griffiths, 2019). JLC, YYL, YS, YPB, and JS prepared the manuscript. All authors checked
Systematic screening tools for digital addiction and various subtypes are the study results and revised the manuscript to the final version.
needed in future studies and clinical practice.

Declaration of Competing Interest

4.4. Limitations

The authors declare no conflict of interest.

As far as we know, this is the first meta-analysis that analyzed and
compared the prevalence characteristics of multiple subtypes of digital
addiction together, and covered the largest population among studies on Appendix A. Supplementary data
similar topics. However, several limitations need to be addressed.
Firstly, considerable heterogeneity was identified across most estimates Supplementary data to this article can be found online at https://doi.
(I2 > 75%), creating large uncertainty around the pooled estimates. org/10.1016/j.cpr.2022.102128.
Through subgroup analysis and meta-regression, this heterogeneity
could be partly explained by the variance of study design (e.g., WHO/ References
World Bank region, publication period, gender, and sample represen-
Al-Khani, A. M., Saquib, J., Rajab, A. M., Khalifa, M. A., Almazrou, A., & Saquib, N.
tativeness). It needs to be emphasized that, as discussed above, differ- (2021). Internet addiction in gulf countries: A systematic review and meta-analysis.
ences in assessment scales for each subtype of digital addiction increased Journal of Behavioral Addictions, 10, 601–610. https://doi.org/10.1556/
the heterogeneity. Secondly, to ensure the quality of included data, the 2006.2021.00057
Alhassan, A. A., Alqadhib, E. M., Taha, N. W., Alahmari, R. A., Salam, M., &
analysis relied on studies published in peer-reviewed journals and did Almutairi, A. F. (2018). The relationship between addiction to smartphone usage
not include grey literatures. This strategy may avoid increasing het- and depression among adults: A cross sectional study. BMC Psychiatry, 18, 148.
erogeneity but was also likely to result in miss of potentially available https://doi.org/10.1186/s12888-018-1745-4
Almourad, M. B., McAlaney, J., Skinner, T., Pleya, M., & Ali, R. (2020). Defining digital
data. Thirdly, some included studies adapted online sampling, volunteer
addiction: Key features from the literature. Psihologija, 53, 237–253.
participation, or convenient sampling, which increased the risk to get Asrese, K., & Muche, H. (2020). Online activities as risk factors for problematic internet
selective samples. To reduce the risk of bias caused by selective sam- use among students in Bahir Dar University, north West Ethiopia: A hierarchical
regression model. PLoS One, 15, Article e0238804. https://doi.org/10.1371/journal.
pling, we had already excluded some high-risk studies through NOS
pone.0238804
quality assessment. Fourthly, because of the absence of data, the results Bell, V., Bishop, D. V., & Przybylski, A. K. (2015). The debate over digital technology and
of this study were unable to fully represent the conditions of low-income young people. British Medical Journal, 351, Article h3064. https://doi.org/10.1136/
countries. Finally, compared with clinical interviews, the self-report bmj.h3064
Buneviciene, I., & Bunevicius, A. (2020). Prevalence of Internet addiction in healthcare
scales applied in most of the included studies were more likely to get professionals: Systematic review and meta-analysis. International Journal of Social
results deviated from the actual conditions. Psychiatry. , Article 0020764020959093. https://doi.org/10.1177/
0020764020959093
Cheng, C., & Li, A. Y. (2014). Internet addiction prevalence and quality of (real) life: A
4.5. Conclusions meta-analysis of 31 nations across seven world regions. Cyberpsychology, Behavior
and Social Networking, 17, 755–760. https://doi.org/10.1089/cyber.2014.0317
In conclusion, by synthesizing the prevalence data, we found that up Chia, D. X. Y., Ng, C. W. L., Kandasami, G., Seow, M. Y. L., Choo, C. C., Chew, P. K. H.,
Lee, C., & Zhang, M. W. B. (2020). Prevalence of internet addiction and gaming
to one-fourth of the general population could be affected by at least one disorders in Southeast Asia: A meta-analysis. International Journal of Environmental
subtype of digital addiction. Prevalence ranged from 6.04% for game Research and Public Health, 17, 2582. https://doi.org/10.3390/ijerph17072582
addiction to 26.99% for smartphone addiction, with big variations be- Christakis, D. A. (2019). The challenges of defining and studying "digital addiction" in
children. JAMA, 321, 2277–2278. https://doi.org/10.1001/jama.2019.4690
tween regions, economic levels, time periods of publication, genders,
Chung, T., Sum, S., Chan, M., Lai, E., & Cheng, N. (2019). Will esports result in a higher
and assessment scales. Given the widespread use of digital media and its prevalence of problematic gaming? A review of the global situation. Journal of
potential adverse effects on physical and mental health, this study could Behavioral Addictions, 8, 384–394. https://doi.org/10.1556/2006.8.2019.46
Dahl, D., & Bergmark, K. H. (2020). Persistence in problematic internet use-a systematic
help to understand the characteristics of different subtypes of digital
review and meta-analysis. Frontiers in Sociology, 5, 30. https://doi.org/10.3389/
addiction, and provide evidence for developing population-specific and fsoc.2020.00030
cultural-specific preventive management and intervention strategies. An Deeks, J. J., Higgins, J. P., & Altman, D. (2019). Analysing data and undertaking meta-
ongoing challenge in the field is that standard criteria for digital analyses. In Cochrane Handbook for Systematic Reviews of Interventions (pp. 241–284).
Dong, H., Yang, F., Lu, X., & Hao, W. (2020). Internet addiction and related psychological
addiction have not been established. Greater agreement in terms of the factors among children and adolescents in China during the coronavirus disease
definitions, subtypes, and measures of digital addiction is needed. 2019 (COVID-19) epidemic. Frontiers in Psychiatry, 11, Article 00751. https://doi.
org/10.3389/fpsyt.2020.00751
Egger, M., Davey Smith, G., Schneider, M., & Minder, C. (1997). Bias in meta-analysis
detected by a simple, graphical test. British Medical Association, 315, 629–634.
https://doi.org/10.1136/bmj.315.7109.629

10
S.-Q. Meng et al. Clinical Psychology Review 92 (2022) 102128

Fam, J. Y. (2018). Prevalence of internet gaming disorder in adolescents: A meta-analysis Mamun, M. A. A., & Griffiths, M. D. (2019). The association between Facebook addiction
across three decades. Scandinavian Journal of Psychology, 59, 524–531. https://doi. and depression: A pilot survey study among Bangladeshi students. Psychiatry
org/10.1111/sjop.12459 Research, 271, 628–633. https://doi.org/10.1016/j.psychres.2018.12.039
Feng, W., Ramo, D. E., Chan, S. R., & Bourgeois, J. A. (2017). Internet gaming disorder: Mboya, I. B., Leyaro, B. J., Kongo, A., Mkombe, C., Kyando, E., & George, J. (2020).
Trends in prevalence 1998–2016. Addictive Behaviors, 75, 17–24. https://doi.org/ Internet addiction and associated factors among medical and allied health sciences
10.1016/j.addbeh.2017.06.010 students in northern Tanzania: A cross-sectional study. BMC Psychology, 8, 73.
Ferguson, C. J., Coulson, M., & Barnett, J. (2011). A meta-analysis of pathological https://doi.org/10.1186/s40359-020-00439-9
gaming prevalence and comorbidity with mental health, academic and social Meng, S., Dong, P., Sun, Y., Li, Y., Chang, X., Sun, G., Zheng, X., Sun, Y., Sun, Y.,
problems. Journal of Psychiatric Research, 45, 1573–1578. https://doi.org/10.1016/j. Yuan, K., Sun, H., Wang, Y., Zhao, M., Tao, R., Domingo, C., Bao, Y., Kosten, T.,
jpsychires.2011.09.005 Lu, L., & Shi, J. (2020). Guidelines for prevention and treatment of internet addiction
Grusser, S. M., Thalemann, R., Albrecht, U., & Thalemann, C. N. (2005). Excessive in adolescents during home quarantine for the COVID-19 pandemic. Heart and Mind,
computer usage in adolescents-results of a psychometric evaluation. Wiener Klinische 4, 95–99. https://doi.org/10.4103/hm.hm_36_20
Wochenschrift, 117, 188–195. https://doi.org/10.1007/s00508-005-0339-6 Mihara, S., & Higuchi, S. (2017). Cross-sectional and longitudinal epidemiological
Grusser, S. M., Thalemann, R., & Griffiths, M. D. (2007). Excessive computer game studies of internet gaming disorder: A systematic review of the literature. Psychiatry
playing: Evidence for addiction and aggression? Cyberpsychology and Behavior, 10, and Clinical Neurosciences, 71, 425–444. https://doi.org/10.1111/pcn.12532
290–292. https://doi.org/10.1089/cpb.2006.9956 Modara, F., Rezaee-Nour, J., Sayehmiri, N., Maleki, F., Aghakhani, N., Sayehmiri, K., &
Ha, J., Yu, J. H., Park, D. H., Ryu, S. H., & Kim, S. J. (2012). Usefulness of Young’s Rezaei-Tavirani, M. (2017). Prevalence of internet addiction in Iran: A systematic
internet addiction test for clinical populations. European Neuropsychopharmacology, review and meta-analysis. Addiction and Health, 9, 243–252.
22, S410. Moher, D., Liberati, A., Tetzlaff, J., & Altman, D. (2009). Preferred reporting items for
Hawi, N. S., Samaha, M., & Griffiths, M. D. (2019). The digital addiction scale for systematic reviews and meta-analyses: The PRISMA statement. PLoS Medicine, 6,
children: Development and validation. Cyberpsychology, Behavior and Social Article e1000097.
Networking, 22, 771–778. https://doi.org/10.1089/cyber.2019.0132 Moreno, M. A., Jelenchick, L., Cox, E., Young, H., & Christakis, D. A. (2011). Problematic
Higgins, J. P., & Thompson, S. G. (2002). Quantifying heterogeneity in a meta-analysis. internet use among US youth: A systematic review. Archives of Pediatrics and
Statistics in Medicine, 21, 1539–1558. https://doi.org/10.1002/sim.1186 Adolescent Medicine, 165, 797–805. https://doi.org/10.1001/archpediatrics.2011.58
Higgins, J. P., Thompson, S. G., Deeks, J. J., & Altman, D. G. (2003). Measuring Pan, Y. C., Chiu, Y. C., & Lin, Y. H. (2020). Systematic review and meta-analysis of
inconsistency in meta-analyses. British Medical Journal, 327, 557–560. https://doi. epidemiology of internet addiction. Neuroscience and Biobehavioral Reviews, 118,
org/10.1136/bmj.327.7414.557 612–622. https://doi.org/10.1016/j.neubiorev.2020.08.013
Humphreys, G. (2019). Sharpening the focus on gaming disorder. Bulletin of the World Paulus, F. W., Ohmann, S., von Gontard, A., & Popow, C. (2018). Internet gaming
Health Organization, 97, 382–383. https://doi.org/10.2471/BLT.19.020619 disorder in children and adolescents: A systematic review. Developmental Medicine
Islam, M. S., Sujan, M. S. H., Tasnim, R., Ferdous, M. Z., Masud, J. H. B., Kundu, S., and Child Neurology, 60, 645–659. https://doi.org/10.1111/dmcn.13754
Mosaddek, A. S. M., Choudhuri, M. S. K., Kircaburun, K., & Griffiths, M. D. (2020). Potenza, M. N., Higuchi, S., & Brand, M. (2018). Call for research into a wider range of
Problematic internet use among young and adult population in Bangladesh: behavioural addictions. Nature, 555, 30. https://doi.org/10.1038/d41586-018-
Correlates with lifestyle and online activities during the COVID-19 pandemic. 02568-z
Addictive Behaviors Reports, 12, Article 100311. https://doi.org/10.1016/j. Sahu, M., Gandhi, S., & Sharma, M. K. (2019). Mobile phone addiction among children
abrep.2020.100311 and adolescents: A systematic review. Journal of Addictions Nursing, 30, 261–268.
Jackson, D., Law, M., Rücker, G., & Schwarzer, G. (2017). The hartung-knapp https://doi.org/10.1097/jan.0000000000000309
modification for random-effects meta-analysis: A useful refinement but are there any Shao, Y. J., Zheng, T., Wang, Y. Q., Liu, L., Chen, Y., & Yao, Y. S. (2018). Internet
residual concerns? Statistics in Medicine, 36, 3923–3934. https://doi.org/10.1002/ addiction detection rate among college students in the People’s Republic of China: A
sim.7411 meta-analysis. Child and Adolescent Psychiatry and Mental Health, 12, 25. https://doi.
Jahan, I., Hosen, I., Al Mamun, F., Kaggwa, M. M., Griffiths, M. D., & Mamun, M. A. org/10.1186/s13034-018-0231-6
(2021). How has the COVID-19 pandemic impacted internet use behaviors and Siste, K., Hanafi, E., Sen, L. T., Christian, H., Adrian, Siswidiani, L. P., Limawan, A. P.,
facilitated problematic internet Use? A Bangladeshi study. Psychology Research and Murtani, B. J., & Suwartono, C. (2020). The impact of physical distancing and
Behavior Management, 14, 1127–1138. https://doi.org/10.2147/prbm.S323570 associated factors towards internet addiction among adults in Indonesia during
Jang, M. H., & Ji, E. S. (2012). Gender differences in associations between parental COVID-19 pandemic: A nationwide web-based study. Frontiers in Psychiatry, 11,
problem drinking and early adolescents’ internet addiction. Journal for Specialists in Article 580977. https://doi.org/10.3389/fpsyt.2020.580977
Pediatric Nursing, 17, 288–300. https://doi.org/10.1111/j.1744-6155.2012.00344.x Sohn, S., Rees, P., Wildridge, B., Kalk, N. J., & Carter, B. (2019). Prevalence of
Khumsri, J., Yingyeun, R., Manwong, M., Hanprathet, N., & Phanasathit, M. (2015). problematic smartphone usage and associated mental health outcomes amongst
Prevalence of Facebook addiction and related factors among Thai high school children and young people: A systematic review, meta-analysis and GRADE of the
students. Journal of the Medical Association of Thailand, 98, S51–S60. evidence. BMC Psychiatry, 19, 356. https://doi.org/10.1186/s12888-019-2350-x
Király, O., Potenza, M. N., Stein, D. J., King, D. L., Hodgins, D. C., Saunders, J. B., Stang, A. (2010). Critical evaluation of the Newcastle-Ottawa scale for the assessment of
Griffiths, M. D., Gjoneska, B., Billieux, J., Brand, M., Abbott, M. W., the quality of nonrandomized studies in meta-analyses. European Journal of
Chamberlain, S. R., Corazza, O., Burkauskas, J., Sales, C. M. D., Montag, C., Epidemiology, 25, 603–605. https://doi.org/10.1007/s10654-010-9491-z
Lochner, C., Grünblatt, E., Wegmann, E., … Demetrovics, Z. (2020). Preventing Statista. (2021). Global digital population as of January 2021. Available from: https
problematic internet use during the COVID-19 pandemic: Consensus guidance. ://www.statista.com/statistics/617136/digital-population-worldwide/. Last
Comprehensive Psychiatry, 100, Article 152180. https://doi.org/10.1016/j. accessed on Nov 19 2021.
comppsych.2020.152180 Sterne, J. A., & Egger, M. (2001). Funnel plots for detecting bias in meta-analysis:
Kuss, D. J., Griffiths, M. D., Karila, L., & Billieux, J. (2014). Internet addiction: A Guidelines on choice of axis. Journal of Clinical Epidemiology, 54, 1046–1055.
systematic review of epidemiological research for the last decade. Current https://doi.org/10.1016/s0895-4356(01)00377-8
Pharmaceutical Design, 20, 4026–4052. https://doi.org/10.2174/ Stevens, M. W. R., Dorstyn, D., Delfabbro, P. H., & King, D. L. (2020). Global prevalence
13816128113199990617 of gaming disorder: A systematic review and meta-analysis. Australian & New
Li, H. H., Wang, L., & Wang, J. Q. (2008). The difference of mental health levels and Zealand Journal of Psychiatry, 0004867420962851. https://doi.org/10.1177/
personality traits between internet social addiction and internet game addiction in 0004867420962851
college students. Chinese Journal of Clinical Psychology, 16, 413–416. Su, W., Han, X., Jin, C., Yan, Y., & Potenza, M. N. (2019). Are males more likely to be
Li, L., Xu, D. D., Chai, J. X., Wang, D., Li, L., Zhang, L., Lu, L., Ng, C. H., Ungvari, G. S., addicted to the internet than females? A meta-analysis involving 34 global
Mei, S. L., & Xiang, Y. T. (2018). Prevalence of internet addiction disorder in Chinese jurisdictions. Computers in Human Behavior, 99, 86–100. https://doi.org/10.1016/j.
university students: A comprehensive meta-analysis of observational studies. Journal chb.2019.04.021
of Behavioral Addictions, 7, 610–623. https://doi.org/10.1556/2006.7.2018.53 Sun, Y., Li, Y., Bao, Y., Meng, S., Sun, Y., Schumann, G., Kosten, T., Strang, J., Lu, L., &
Li, Y. Y., Sun, Y., Meng, S. Q., Bao, Y. P., Cheng, J. L., Chang, X. W., Ran, M. S., Sun, Y. K., Shi, J. (2020). Brief report: Increased addictive internet and substance use behavior
Kosten, T., Strang, J., Lu, L., & Shi, J. (2021). Internet addiction increases in the during the COVID-19 pandemic in China. The American Journal on Addictions, 29,
general population during COVID-19: Evidence from China. The American Journal on 268–270. https://doi.org/10.1111/ajad.13066
Addictions, 30, 389–397. https://doi.org/10.1111/ajad.13156 Tiego, J., Lochner, C., Ioannidis, K., Brand, M., Stein, D. J., Yucel, M., Grant, J. E., &
Lin, M. P. (2020). Prevalence of internet addiction during the COVID-19 outbreak and its Chamberlain, S. R. (2019). Problematic use of the internet is a unidimensional quasi-
risk factors among junior high school students in Taiwan. International Journal of trait with impulsive and compulsive subtypes. BMC Psychiatry, 19, 348. https://doi.
Environmental Research and Public Health, 17. https://doi.org/10.3390/ org/10.1186/s12888-019-2352-8
ijerph17228547 Wang, Y., Shi, L., Que, J., Lu, Q., Liu, L., Lu, Z., Xu, Y., Liu, J., Sun, Y., Meng, S., Yuan, K.,
Lipsey, M., & Wilson, D. B. (2000). Practical meta-analysis. Los Angeles: SAGE Ran, M., Lu, L., Bao, Y., & Shi, J. (2021). The impact of quarantine on mental health
Publication. Inc. status among general population in China during the COVID-19 pandemic. Molecular
Long, J., Liu, T., Liu, Y., Hao, W., Maurage, P., & Billieux, J. (2018). Prevalence and Psychiatry, 26, 4813–4822. https://doi.org/10.1038/s41380-021-01019-y
correlates of problematic online gaming: A systematic review of the evidence Weinstein, A., & Lejoyeux, M. (2010). Internet addiction or excessive internet use.
published in Chinese. Current Addiction Reports, 5, 359–371. https://doi.org/ American Journal of Drug and Alcohol Abuse, 36, 277–283. https://doi.org/10.3109/
10.1007/s40429-018-0219-6 00952990.2010.491880
Lopez-Fernandez, O. (2018). Generalised versus specific internet use-related addiction WHO. (2014). Public health implications of excessive use of the internet, computers,
problems: A mixed methods study on internet, gaming, and social networking smartphones and similar electronic devices: Meeting report. Available from. Last
behaviours. International Journal of Environmental Research and Public Health, 15, accessed on Nov 27 2021 https://www.who.int/publications/i/item/9789241509
2913. https://doi.org/10.3390/ijerph15122913 367.

11
S.-Q. Meng et al. Clinical Psychology Review 92 (2022) 102128

WHO. (2020). Healthy at home – Mental health. Available from. Last accessed on Nov 27 World Bank. (2021). World Bank country and lending groups. Available from. Last accessed
2021 https://www.who.int/campaigns/connecting-the-world-to-combat-coronavi on Nov 27 2021 https://datahelpdesk.worldbank.org/knowledgebase/arti
rus/healthyathome/healthyathome—mental-health. cles/906519-world-bank-country-and-lending-groups.
WHO. (2021). Mental Health Atlas 2020. Available from. Last accessed on Nov 27 2021 Zhang, M. W. B., Lim, R. B. C., Lee, C., & Ho, R. C. M. (2018). Prevalence of internet
https://www.who.int/publications/i/item/9789240036703. addiction in medical students: A meta-analysis. Academic Psychiatry, 42, 88–93.
WHO. (2021). WHO regional offices. Available from. Last accessed on Nov 27 2021 http https://doi.org/10.1007/s40596-017-0794-1
s://www.who.int/about/who-we-are/regional-offices.

12