Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Nov 1.
Published in final edited form as: Anxiety Stress Coping. 2024 Jul 31;37(6):775–793. doi: 10.1080/10615806.2024.2383766

Rates and psychological stress predictors of problematic internet use (PIU) during the COVID-19 pandemic in a racially diverse sample of young adults

G Alice Woolverton 1, Courtney Stevens 1,2, Hyeouk Chris Hahm 3, Cindy H Liu 1,4,5
PMCID: PMC11489012  NIHMSID: NIHMS2015196  PMID: 39085999

Abstract

Background

Problematic internet use (PIU), which includes social media misuse (SMM) and gaming misuse (GM), is uncontrollable and associated with significant psychological impairment. PIU is a coping behavior for COVID-19 related stress. We explored distress-related predictors of PIU in a young adult racially diverse sample during the pandemic.

Methods

Analyses used cross-sectional survey data (N = 1,956). Psychological diagnoses, financial distress, COVID-19-related emotions, psychological distress, distress tolerance, social support, loneliness, SMM and GM were measured. Hierarchical multiple regressions identified predictors of PIU. Race-stratified exploratory analyses sought to understand if predictors held true across racial groups.

Results

Low distress tolerance was associated with SMM and GM, as were depression symptoms, with racial differences observed. SMM was associated with younger age, and GM was associated with male gender. PTSD symptoms predicted more GM. SMM and GM rates varied between racial groups. COVID-19-related adjustment challenges and stress predicted SMM and GM respectively, with racial differences observed.

Conclusion

Individual psychological distress and low distress tolerance markedly increased PIU risk. Clinicians should screen for stress-related PIU risk factors and bolster distress tolerance in vulnerable patients. Comparing PIU to different forms of coping in a larger sample would further clarify groups differences in stress coping behaviors.

Keywords: problematic internet use, psychological stress, COVID-19 pandemic, racial differences, social media misuse, gaming misuse

1. Introduction

Problematic internet use (PIU) is a behavioral addiction characterized by obsessive and uncontrollable internet use with associated withdrawal symptoms caused by lack of use (Demetrovics et al., 2016; Shapira et al., 2000). Significant detrimental consequences are associated with PIU, including neglect of activities of daily living, work, school, and/or relationships, and functional impairments such as altering sleep and eating in a determinantal way (Spada, 2014). Determining predictors and risk factors associated with PIU is critical to help clarify the disorder’s etiology and lead to early detection and/or prevention of the disorder. Early detection and prevention are paramount, as PIU is associated with lower self-esteem and life satisfaction over time (Kraut et al., 1998; Ranjan et al., 2021; Xiong et al., 2023). PIU can be generalized, which refers to a “multidimensional overuse of the internet itself” (Caplan, 2002, p. 556), or specific, which is a discrete addictive behavior accessed through the internet (e.g., gaming misuse, also known as Internet Gaming Disorder (IGD); Laconi et al., 2015; Lopez-Fernandez, 2018). Since The American Psychiatric Association (APA) is only considering the addition of a specific PIU disorder, IGD, to its next Diagnostic and Statistical Manual (APA, 2013), we focus our study on specific PIU.

Our paper focuses on two common forms of specific PIU that were elevated during the COVID-19 pandemic, social media misuse (SMM) and gaming misuse (GM), which we refer to collectively as PIU (Gómez-Galán et al., 2020; Masaeli et al., 2021). During the pandemic, lockdowns occurred worldwide to prevent the spread of illness, with increasingly distressing news about the morbidity and mortality associated with the virus proliferating through media outlets and online. During this unprecedented time, individuals faced social isolation, with evidence suggesting that loneliness was associated with elevated PIU during COVID-19 lockdown in non-U.S. samples (Alheneidi et al., 2021; Moretta et al., 2020). There is mixed evidence about shifts in social support during this time (e.g. Mitchell et al., 2022); however, those who reported strong social support during the pandemic were less likely to report increased depression, with opportunity remaining to investigate if social support confers similar protection against PIU (Grey et al., 2020). High levels of uncertainty and fear about illness and the future (i.e., COVID-19 related stress and adjustment challenges) were widespread, and PIU has been identified as a behavior to cope with this elevated distress in non-U.S. samples (Xu et al., 2021; Zhao & Zhou, 2021). At the same time, the pandemic brought sudden financial distress to many Americans who faced job loss due to lockdown and overall contraction of economic activity (Lund et al., 2020). While financial distress was positively associated with PIU prior to the pandemic (Aboujaoude, 2010), recent research has yet to identify COVID-19-related financial distress to PIU.

The stressful escapism model of PIU is a prominent framework used to understand the motivations of those who exhibited SMM and GM during the pandemic (e.g., Fernandes et a., 2020; Singh et al., 2022). In this model, individuals engage in escapist behaviors to distract from overwhelming environmental stress (Snodgrass et al., 2014). Using this theory, being highly stressed and needing to escape, feelings that were likely elevated during the COVID-19 pandemic, motivates individuals to engage in more PIU (Jarman et al., 2021; Stenseng et al. 2021; Yan, et al., 2014). For example, in a qualitative study examining PIU, one participant noted that when life was particularly stressful and overwhelming, it became increasingly difficult to stop playing online videogames (Snodgrass et al., 2014). One factor that may protect against escapist coping with PIU is high distress tolerance, which reflects one’s ability to tolerate negative emotional experiences without compensatory escapism behaviors. Distress tolerance is protective against various forms of addictive behaviors such as substance use disorder and binge eating (e.g., Kozak & Fought, 2011; Son & Geong, 2023). Prior to the COVID-19 pandemic, a few studies identified a link between distress tolerance and PIU (e.g., Elhai et al., 2018; Skues et al., 2016), with opportunity remaining to explore this potential link during the pandemic.

PIU also has known mental health correlates, such as positive associations with anxiety and depression (Bányai et al., 2017; Stevens et al., 2020; Tham et al., 2020). PIU and depression symptoms are bidirectionally related, with depression symptoms both contributing to increased PIU as a method to cope with negative emotions, and exacerbating depressive symptoms (Dillman-Carpentier et al., 2008; Horwitz et al., 2011; Romer et al., 2013; Wartberg et al., 2018). With respect to the relationship between PIU and particular anxiety disorders, PIU is positively associated with Social Anxiety Disorder by contributing to avoidance of, and increased worry about, face-to-face social interactions (Caplan, 2006; Lee & Stapinski, 2012). Interestingly, although Generalized Anxiety Disorder (GAD) is the most commonly diagnosed anxiety disorder, there is limited research examining GAD’s potential relationship with PIU. Similarly, despite a well-documented pattern of escape and avoidance behaviors associated with Post-Traumatic Stress Disorder (PTSD), no known research to date has examined a potential relationship with PIU (Charlton & Thompson, 1996).

Drawing from the stressful escapism model of PIU (Snodgrass et al., 2014), the primary research goal of this study was to identify factors that increased vulnerability to PIU during the COVID-19 pandemic. Factors examined in this study include: social experiences (i.e., loneliness and social support), pandemic-related experiences (i.e., COVID-19 stress, COVID-19 adjustment challenges), stressors that were elevated during the pandemic (i.e., financial stress), co-morbid diagnoses (i.e., depression, anxiety, and PTSD), adverse mental health symptoms (i.e., depression, anxiety, and PTSD symptoms), and distress tolerance. Based on existing evidence demonstrating the relationship between mental disorders and PIU, we examine whether external and social stressors are associated with PIU above and beyond psychological diagnoses and symptoms (i.e., while controlling for mental diagnoses and symptoms). We hypothesize that preexisting depression, anxiety, and PTSD, as well as symptoms of these disorders, are associated with more PIU. As PIU is seen as a coping behavior in response to stress as drawn from the stressful escapism model of PIU, we hypothesize that stressful experiences that were elevated during the pandemic for young adults – low social support, distress associated with COVID-19, and financial distress – will predict greater PIU (Park et al., 2020; Refaeli & Achdut, 2021). Relatedly, we hypothesize that low distress tolerance predicts more PIU.

Second, based on emergent research identifying racial differences in PIU rates (e.g., Pew Research Center, 2022), a secondary exploratory research goal examines whether or not PIU risk factors vary between Latinx, Black, Asian, and White participants. White individuals have been found to engage in less PIU than those from other racial groups (e.g., Nagata et al., 2022). Black and Latinx adolescents are more likely to report using the internet “almost constantly” compared to their White peers (Pew Research Center, 2022). Among U.S. high school students, Asian and Latinx students are more likely to meet criteria for PIU than their Black and White peers (Liu et al., 2011). In another adolescent sample, those of color reported higher social media misuse than their White peers (Nagata et al., 2022). Furthermore, in a young adult sample, Asian participants reported higher rates of problematic internet use than Latinx, White, and Black participants (Yates et al., 2012).

Although this emerging work documents racial differences in PIU rates, most studies only compare PIU between White vs. non-White participants with models that do not stratify by race; as such, predictors underlying observed racial differences in PIU are yet to be identified. Stratifying by race to examine patterns of predictors of PIU is needed to determine whether individuals from certain racial groups vary in the specific vulnerabilities to PIU, including certain conditions (i.e., in the context of different stressors or symptoms) or coping with negative emotions. Our study expands on the limited current research, as our stratified models aim to provide exploratory information on whether certain predictors of PIU might be especially strong for some racial ethnic minority groups. Such information could have clinical implications for assessing individuals at risk for PIU who are experiencing certain stressors. Taken together, based on existing limited literature, we hypothesize observing higher rates of PIU among non-White participants in our sample and adopt an exploratory approach to identifying potential differences in predictor variables of PIU.

2. Methods

2.1. Participants

Table 1 outlines the demographic information for the sample. Participants were 1,976 men and women with an average age of 26.05 years (SD=3.14). The majority of participants identified as female and heterosexual. 30.8% of the sample were Asian/Asian American, 22.4% were White, 19.0% were Hispanic/Latinx, and 18.8% were Black/African American. 9.0% were classified as Other, which grouped together the remaining smaller numbers of participants who identified as American Indian or Alaska Native, Middle Eastern or North African, Native Hawaiian or Pacific islander, Multiracial, or another race. Participants reported previous mental health diagnoses of Major Depressive Disorder (29.5%), Generalized Anxiety Disorder (24.1%), and PTSD (9.7%). Current symptoms of depression, anxiety, and PTSD, as well as mean rates of COVID-19 adjustment challenges and COVID-19 stress, financial distress, loneliness, social support, distress tolerance, and social media and GM are reported in Table 1. According to criteria created by GM and SSM measure developers (Van den Eijnden et al., 2016), 11.5% of participants were classified as disordered social media users, while 9.3% of the sample were classified as disordered gaming users.

Table 1:

Sociodemographic characteristics for the full analytic sample (N = 1,956)

Characteristics Mean or % Range
Age (years) 26.05 19.78–32.99
Gender Identity
 Woman 71.90%
 Man 24.90%
 Transgender or self-identify 3.30%
Racial Identity
 Hispanic or Latinx 19.00%
 Black of African American 18.80%
 Asian or Asian American 30.80%
 White 22.40%
 Other + 9.00%
Sexual Orientation
 Straight 70.70%
 Asexual, lesbian, gay, or bisexual 22.70%
 Questioning, self-identify, or, prefer not to answer 6.60%
Depression Diagnosis (Major Depressive Disorder) 29.50%
Anxiety Diagnosis (Generalized Anxiety Disorder) 24.10%
Post-traumatic Stress Disorder Diagnosis (PTSD) 9.70%
Depression Symptoms (PHQ-8) 8.3 0.00–24.00
Anxiety Symptoms (GAD-7) 7.54 000–21.00
PTSD Symptoms (PCL-C) 35.92 17.00–85.00
COVID-19 Related adjustment challenges 20.47 7.00–35.00
COVID-19 Related Stress 3.47 1.00–7.00
Financial Distress 5.48 2.00–8.00
Loneliness 5.61 3.00–9.00
Social Support 5.24 1.00–7.00
Distress Tolerance (DTS Global Score) 3.16 1.00–5.00
Social Media Misuse (SMM) 10.4 9.00–18.00
Percentage meeting criteria for disordered social media user 11.50%
Gaming Misuse (GM) 11.12 9.00–18.00
Percentage meeting criteria for disordered gaming user 9.30% 9.00–18.00
Open in a new tab
+

Other race includes: American Indian or Alaska Native, Middle Eastern or North African, Native Hawaiian or Pacific Islander, Multiracial, or some other race, ethnicity, or origin

2.2. Procedure and Ethical Considerations

Data were obtained from the longitudinal COVID-19 Adult Resilience Experiences Study (CARES), which examined the psychosocial experiences of U.S. young adults aged 18–30 years. Participants were recruited via email listservs, social media, and word of mouth. Surveys were only offered in English, making English proficiency a requirement for participation. Participants who provided informed consent participated in three, 15-to-30-minute online surveys from April 2020-May 2021. The current study examined cross-sectional data from Wave IV (February 2022 - June 2022). Analyses focus on participants who completed the study (N = 1,956). This study was approved by the Institutional Review Board at Boston University and was identified as conferring minimal risk to participants.

2.3. Measures

2.3.1. Sociodemographic Variables.

Age, measured in years, was normally distributed across the overall sample and the four racial groups analyzed separately. In response to the question, “what is your gender?” participants selected woman, man (reference group), trans woman, trans man, or self-identify. In response to “what is your sexual orientation?” participants selected straight/ heterosexual (reference group), asexual, bisexual, gay, lesbian (grouped together for analysis), questioning, prefer not to answer, or self-identify (grouped together for analysis).

2.3.2. Pre-existing Mental Health Diagnoses.

Pre-existing depression, anxiety, and posttraumatic stress disorder (PTSD) diagnoses were assessed through three questions, “have you ever been diagnosed with the following?” Participants rated their answers on a four-point Likert scale of no = 0, suspected but not diagnosed = 1, yes diagnosed but not treated = 2, and yes diagnosed and treated = 3. Responses were re-coded to be binary, with confirmed diagnoses (2 and 3) coded as having a diagnosis, and responses 0 and 1 coded as not having a diagnosis.

2.3.3. Depression Symptoms.

Participants completed the eight-item Patient Health Questionnaire (PHQ-8) to assess for past two-week symptoms of Major Depressive Disorder (Pressler et al., 2011). Responses range on a four-point Likert scale from 0 = not at all to 3 = nearly every day. Items are summed to provide a total score that may range from 0 to 24, with higher scores indicating greater depressive symptoms. The reliability of the PHQ-8 in the current sample was excellent (α=0.913).

2.3.4. Anxiety Symptoms.

Participants completed the seven-item General Anxiety Disorder-7 (GAD-7; Spitzer et al., 2006), which detects symptoms consistent with GAD across diverse populations. Responses range on a four-point Likert scale ranging from 0 = not at all to 3 = nearly every day. Items are summed to provide a total score that may range from 0–21, with higher scores indicating greater anxiety symptoms. Reliability calculated from our data indicate excellent internal consistency (α=0.935).

2.3.5. PTSD Symptoms.

Participants completed the Post-traumatic Stress Disorder Checklist: Civilian Scale (PCL-C; Andrykowski et al., 1998), which assesses clinical symptoms associated with PTSD. The PCL-C consists of 17 items, with responses ranging on a five-point Likert scale from 1 = not at all to 5 = extremely. Items are summed to provide a total score that may range from 17–85, with higher scores indicating greater PTSD symptoms. Reliability calculated from our data indicate excellent internal consistency (α=0.958).

2.3.6. COVID-19 Stress.

Participants reported COVID-19 stress in response to, “please indicate the overall level of stress in the past month due to the COVID-19 pandemic.” Responses ranged on a seven-point Likert scale (1 = not at all, 2 = low, 3 = slightly, 4 = somewhat, 5 = moderately, 6 = very, 7 = extremely). Higher scores indicated higher levels of COVID-19 stress.

2.3.7. COVID-19 Adjustment Challenges.

Trouble adapting and feeling worried, dazed, empty, or bitter about the pandemic were measured with seven questions, each of which ranged on a five-point Likert scale from 1 = strongly disagree to 5 = strongly agree. Items were summed to provide a total score that may range from 7 to 35, with higher scores indicating greater COVID-19 adjustment challenges. Reliability calculated from our data indicate good internal consistency (α=0.829).

2.3.8. Financial Distress.

Financial distress was assessed with two items: “I feel stressed about my personal finances in general” and “I worry about being able to pay monthly expenses.” Responses varied on a four-point Likert scale that ranged from 1 = strongly disagree to 4 = strongly agree. The two items were summed to provide a total score that may range from 2–8, with higher scores indicating greater financial distress. Reliability calculated from our data indicate adequate internal consistency (α=0.786).

2.3.9. Distress Tolerance.

Participants completed the Distress Tolerance Scale (DTS; Simons and Gaher, 2005) to measure tolerance to and emotional regulation in response to aversive stimuli. The DTS is a fifteen-item measure; responses range on a five-point Likert scale from 5 = strongly disagree to 1 = strongly agree. A mean score was calculated from all items in the scale, with higher scores indicating higher average distress tolerance. Reliability calculated from our data indicate good internal consistency (α=0.892).

2.3.10. Social Support.

The Multidimensional Scale of Perceived Social Support (MSPSS; Zimet et al., 1988) is a twelve-item tool that measures past two-week feelings of social support. Responses ranged on a seven-point Likert scale from 1 = very strongly disagree to 7 = very strongly agree. A mean score was calculated from all items in the scale, with higher scores indicating higher levels of social support. Possible scores ranged from 1–7. Reliability calculated from our data indicate excellent internal consistency (α=0.926).

2.3.11. Loneliness.

The UCLA Three-Item Loneliness Scale (LS-SF; Hughes et al., 2004) assesses past-two-week loneliness. Items assess for lack of companionship, feeling left out, and feeling isolated. Answers range on a three-point scale from 1 = hardly ever to 3 = often. The three items are summed to provide a total score that may range from 3–9, with higher scores indicating greater loneliness. Reliability calculated from our data indicate good internal consistency (α=0.827).

2.3.12. Social Media Misuse (SMM).

Addictive thoughts and behaviors regarding social media use were assessed with The Social Media Disorder Scale- Short Form (SMDS-SF; Van den Eijnden et al., 2016). This scale consists of nine yes/ no (yes = 2, no = 1) questions which mirror the nine diagnostic criteria for the proposed IGD. In other words, the SMDS-SF measures the same nine constructs, such as preoccupation, tolerance, and withdrawal, as IGD (APA, 2013). Items were summed to provide a total score that may range from 9–18, with higher scores indicating greater social media misuse. As recommended by scale developers, endorsing five or more items resulted in classification as a disordered social media user, which aligns with DSM-5’s proposal to diagnose IGD if five or more diagnostic criteria are endorsed (APA, 2013). The SMDS-SF is the first scale in the literature which aligns its items and its clinical cut-off with DSM-5’s proposed criteria for Internet Gaming Disorder (Van den Eijnden et al., 2016). Reliability calculated from our data indicate adequate internal consistency (α=0.773).

2.3.13. Gaming Misuse (GM).

Due to the dearth of validated GM measures and the alignment of the SMDS-SF with DSM-5’s proposed IGD, an adapted version of the SMDS-SF, in which “social media” was replaced with “gaming” for all nine items, was used to measure addictive gaming thoughts and behaviors. Sum scores were used, with higher scores indicating more gaming misuse (GM). As this measure aligns with criteria used to classify disordered social media users and with DSM-5’s proposed criteria for diagnosing IGD, we posit that it has sufficient content validity in measuring gaming misuse. Individuals endorsing five or more items were classified as disordered gamers (Van den Eijnden et al., 2016). Reliability calculated from our data indicate good internal consistency (α=0.83).

2.4. Data Analysis

To identify variables that were related to PIU, hierarchical multiple regressions were conducted, first with SMM and then with GM as the outcome measure. In each multiple regression, the first block included sociodemographic characteristics (age, race, gender, and sexual orientation). Subsequent blocks were added to assess the sequential added effects of predictor variables. In the second block, pre-existing depression, anxiety, and PTSD were entered. In the third block current symptoms of depression, anxiety, and PTSD were entered to examine whether these symptoms were associated with PIU while holding previous diagnoses of these disorders constant. The fourth block included distress variables (i.e. COVID-19 adjustment challenges, COVID-19 stress, and financial distress) to examine whether external stressors were associated with PIU while holding mental health variables in previous blocks constant.

The fifth block included global distress tolerance to examine whether the ability to tolerate negative emotions was associated with PIU while holding mental health and distress variables constant. Finally, to test whether social experiences that may have been impacted by COVID-19 were associated with PIU above and beyond distress and mental health variables, social support and loneliness were entered in the sixth block.

Given our secondary exploratory aim to examine predictors of PIU within racial groups, means testing was conducted to identify group differences between predictor variables. Next, using the same predictor variables as above (omitting race), multiple hierarchical regressions were performed to examine stability or variation in PIU predictor variables between Latinx, Black, Asian, and White groups. The Other racial group included in primary analyses was excluded in stratified models, as we choose to focus on Latinx, Black, Asian, and White racial groups in exploring racial differences.

3. Results

Mean rates of predictor variables examined by race, and percentages of those meeting criteria as disordered users, are presented in Table 2. Racial group differences in predictor variable means were identified for all variables except for COVID-19 related stress. Due to concern for correlation between predictor variables, variance inflation factors (VIFs) were calculated to detect multicollinearity for both SMM and GM hierarchical multiple regressions. VIFs ranged from 1.0 – 2.1, with the majority of values ranging from 1.0 – 3.5. This indicates acceptable correlation between predictor variables to continue with regression analyses, as VIF values greater than 5 are indicative of multicollinearity (Tsagris & Pandis, 2021).

Table 2:

Racial group differences between predictor variables

Predictor Variables Latinx (n=372) Black (n=368) Asian (n=602) White (n=439) F-Ratio p-value
Group Means or %
Social Media Misuse (SMM) 11.09a 10.83ab 10.94ab 10.61b 4.218 0.006*
% Meeting criteria for disordered social media user 15.30% 13.60% 13.10% 5.70%
Gaming Misuse (GM) 11.22a 11.21a 11.34a 10.39b 5.99 <.001*
% Meeting criteria for disordered gaming user 12.60% 18.90% 21.60% 5.20%
Depression Diagnosis (Major Depressive Disorder) 0.35a 0.29ab 0.23b 0.32a 5.55 <.001*
Anxiety Diagnosis (Generalized Anxiety Disorder) 0.28a 0.19b 0.18b 0.3a 9.25 <.001*
Post-traumatic Stress Disorder Diagnosis (PTSD) 0.12ab 0.12a 0.67b 0.07ab 3.72 0.011*
Depression Symptoms (PHQ-8) 9.56a 9.13a 7.55b 7.25b 13.93 <.001*
Anxiety Symptoms (GAD-7) 8.36a 7.73a 6.67b 7.64ab 6.58 <.001*
PTSD Symptoms (PCL-C) 39.34a 38.35a 34.02b 32.47b 17.36 <.001*
COVID-19 Related adjustment challenges 21.1a 19.09b 20.59a 20.97a 9.07 <.001*
COVID-19 Related Stress 3.61 3.34 3.46 3.45 1.7 0.165
Financial Distress 6.03a 5.65b 5.41b 5.04c 23.13 <.001*
Distress Tolerance (DTS Global Score) 2.97a 3.09ab 3.19b 3.37c 15.55 <.001*
Social Support 5.01a 4.81a 5.32b 5.72c 42.8 <.001*
Loneliness 5.83a 5.83a 5.44b 5.34b 7.63 <.001*
Open in a new tab
*

p<.05

Note: Means not sharing subscripts differ at α = .05 indicated by Tukey HSD

3.1.0. Predictors of SMM Across the Full Sample.

Results of hierarchical multiple regression analyses to determine predictors of SMM are displayed in Table 3. The overall model for SMM was significant F(21)=18.71, p < .001, accounting for 16.3% of the variance SMM. Age was negatively associated with SMM (β = −0.105, p < .001), whereas female gender was positively associated with SMM relative to male gender (β = 0.051, p < .05). Compared to White participants, those who identified as Latinx (β = 0.106, p < .001), Black (β = 0.059, p < .05), or Asian (β = 0.090, p < .01) engaged in significantly more SMM. Sexual orientation did not significantly predict SMM. After controlling for sociodemographic variables, preexisting PTSD predicted significantly less SMM (β = −0.059, p < .05); depression and anxiety diagnoses did not predict SMM. Those with more PTSD symptoms engaged in more SMM (β = 0.197, p < .001) than those with fewer symptoms. Anxiety and depression symptoms did not significantly predict SMM. High COVID-19 adjustment challenges (β = 0.121, p < .001) positively predicted SMM. Low distress tolerance (β = −0.105, p < .001) predicted more SMM. COVID-19 stress, financial distress, social support, loneliness did not significantly predict SMM.

Table 3:

Problematic internet use (SMM and GM) in the full analytic sample (N = 1,956)

Social Media Misuse, Total adjusted R2 =.163 Gaming Misuse, Total adjusted R2 =.208
Unstandardized β Standardized β ΔR 2 Unstandardized β Standardized β ΔR 2
1: Sociodemographics 0.042*** 0.031***
Age −0.058 −0.105*** −0.004 −0.006
Race (ref=White)
Latinx 0.542 0.106*** 0.398 0.068
Black 0.302 0.059* 0.494 0.088*
Asian 0.392 0.090** 0.904 0.178***
Other + −0.073 −0.01 0.37 0.044
Gender (ref=men)
Women 0.227 0.051* −0.824 −0.171***
Transgender or Other Gender 0.259 0.02 −0.399 −0.03
Sexual Orientation (ref=straight)
LGBA 0.182 0.038 −0.075 −0.014
Questioning, prefer not to answer, or other 0.237 0.029 0.08 0.009
2: Pre-Existing Diagnoses 0.005* 0.013**
Depression 0.022 0.005 −0.309 −0.06
Anxiety −0.148 −0.031 0.224 0.04
PTSD −0.399 −0.059* 0.072 0.01
3: Psychological Symptoms 0.101*** 0.148***
Depression Symptoms 0.02 0.063 0.075 0.208***
Anxiety Symptoms −0.001 −0.003 −0.023 −0.061
PTSD Symptoms 0.024 0.197*** 0.029 0.209***
4: Distress 0.013*** 0.013**
COVID-19 Adjustment Challenges 0.041 0.121*** 0.016 0.043
COVID-19 Stress −0.016 −0.013 0.124 0.085*
Financial Distress −0.026 −0.023 −0.047 −0.034
5: Distress Tolerance 0.008*** 0.017***
Global Distress Tolerance −0.246 −0.105*** −0.466 −0.168***
6: Social Connectedness 0.001 0.003
Social Support 0.03 0.019 0.063 0.035
Loneliness 0.051 0.049 −0.052 −0.043
Open in a new tab
+

Other race includes= American Indian or Alaska Native, Middle Eastern or North African, Native Hawaiian or Pacific islander, Multiracial, or some other race, ethnicity or origin

*

p < 0.05

**

p < 0.01

***

p < 0.001

3.2.0. Predictors of GM Across the Full Sample.

Results of hierarchical multiple regression analyses to determine predictors of GM are displayed in Table 3. The overall model for GM was significant F(21)=13.62, p < .001, accounting for 20.8% of the variance in GM. Among sociodemographic variables, women engaged in less GM than men (β = −0.171, p < .001). Compared to White and Latinx participants, Black (β = 0.088, p < .05) and Asian participants (β = 0.178, p < .001) engaged in significantly more GM. Age and sexual orientation did not significantly predict GM. After controlling for sociodemographic variables, preexisting mental health diagnoses did not predict GM. Current depression symptoms (β = 0.208, p < .001) and current PTSD symptoms (β = 0.209, p < .001) predicted more GM. Anxiety symptoms did not predict GM. High COVID stress (β = 0.085, p < .05) predicted more GM. Low distress tolerance (β = −0.168, p < .001) predicted more GM. COVID-19 adjustment challenges, financial distress, social support, and loneliness did not significantly predict GM.

3.3.0. Predictors of SMM Stratified by Race.

Exploratory hierarchical multiple regression analyses examined whether or not predictors of SMM were constant or variable between four racial groups (Table 4). The overall models for SMM were significant for all racial groups (Table 4). While younger age predicted more SMM in the overall sample, in the stratified models, this result only held true for Latinx (β = −0.117, p < .05) and White (β = −0.147, p < .01) participants. The overall finding that women engaged in more SMM than men only held true in White participants in the stratified models (β = 0.116, p < .05). While sexual orientation was not a significant predictor in the overall model, stratified models showed that among Black participants, lesbian, gay, bisexual, and asexual individuals engaged in more SMM than straight participants (β = 0.100, p < .05). Aligned with the overall model, preexisting depression and anxiety diagnoses were not significant predictors of SMM in any racial group. Preexisting PTSD significantly predicted less SMM in the overall model, with this only holding true for Latinx participants in the stratified models (β = −0.151, p < .01). More depression symptoms predicting more SMM only held true among Asian participants in the stratified models (β = 0.148, p < .05). PTSD symptoms’ positive association with SMM observed in the overall model held true across all four racial groups. COVID-19 adjustment challenges predicted more SMM in the overall model, holding true among Black (β = 0.173, p < .01) and Asian (β = 0.155, p < .001) participants in the stratified models. Low distress tolerance’s positive association with SMM observed in the overall model held true across Black (β = −0.129, p < .05), Asian (β = −0.114, p < .05), and White (β = −0.181, p < .001) groups, but not Latinx participants in the stratified models.

Table 4:

Predictors of social media misuse (SMM) stratified by race

Latinx n = 372
Total adjusted R2 =.178		 Black n = 368
Total adjusted R2 =.276		 Asian n = 602
Total adjusted R2 =.134		 White n = 439
Total adjusted R2 =.133
Unstandardized B Standardized B ΔR2 Unstandardized B Standardized B ΔR2 Unstandardized B Standardized B ΔR2 Unstandardized B Standardized B ΔR2
1: Sociodemographics 0.072*** 0.053** 0.013 0.040***
Age −0.069 −0.117* −0.039 −0.066 0.025 0.043 −0.07 −0.147**
Gender (ref=men)
Women 0.208 0.045 −0.098 −0.021 0.208 0.045 0.512 0.116*
Transgender or Other Gender 0.028 0.002 0.958 0.055 0.162 0.009 0.348 0.042
Sexual Orientation (ref=straight)
LGBA 0.077 0.016 0.549 0.100* 0.185 0.033 0.229 0.064
Questioning, prefer not to answer, or other 0.58 0.064 −0.006 −0.001 0.483 0.058 0.298 0.051
2: Pre-Existing Diagnoses 0.029* 0.01 0.002 0.11
Depression 0.353 0.078 −0.028 −0.006 −0.239 −0.049 0.132 0.038
Anxiety 0.325 0.069 −0.551 −0.096 −0.113 −0.021 −0.272 −0.077
PTSD Diagnosis −1.008 −0.151** −0.216 −0.031 −0.492 −0.059 0.02 0.185
3: Psychological Symptoms 0.095*** 0.172*** 0.112*** 0.082***
Depression Symptoms 0.026 0.078 0.001 0.002 0.05 0.148* −0.003 −0.01
Anxiety Symptoms −0.013 −0.038 0.022 0.063 −0.026 −0.071 0.022 0.078
PTSD Symptoms 0.03 0.251** 0.035 0.274** 0.022 0.167* 0.02 0.185*
4: Distress 0.015 0.030** 0.021** 0.006
COVID-19 Adjustment Challenges 0.026 0.076 0.057 0.173** 0.058 0.155*** 0.018 0.061
COVID-19 Stress 0.035 0.028 0.018 0.014 −0.085 −0.064 −0.046 −0.043
Financial Distress −0.132 −0.102 −0.042 −0.036 0.031 0.026 −0.024 −0.025
5: Distress Tolerance 0 0.010* 0.010* 0.026***
Global Distress Tolerance −0.008 −0.003 −0.341 −0.129* −0.292 −0.114* −0.333 −0.181***
6: Social Connectedness 0.005 0.001 0.001 0.001
Social Support −0.047 −0.029 0.046 0.031 0.047 0.024 0.001 0
Loneliness 0.086 0.07 0.014 0.012 0.05 0.045 0.043 0.49
Open in a new tab
*

p < 0.05

**

p < 0.01

***

p < 0.001

3.4.0. Predictors of GM Stratified by Race.

Exploratory hierarchical multiple regression analyses explored whether or not predictors of GM were constant or variable between four racial groups (Table 5). The overall models for GM were significant for all racial groups, as reported in Table 4. Men’s increased rates of GM compared to women observed in the overall model only held true across Black (β = −0.155, p < .05) and Asian (β = −0.201, p < .001) participants in the stratified models. Like the overall model, age, sexual orientation, and preexisting mental health diagnoses were not significant predictors of GM in the stratified models. After controlling for sociodemographic variables and past diagnoses, the overall model finding that more depression symptoms predicted more GM only held true among Latinx (β = 0.270, p < .05) and Asian participants (β = 0.378, p < .001) in the stratified models. The overall model finding that more PTSD symptoms predicted more GM only held true among Latinx participants (β = 0.370, p < .01) in the stratified models. High COVID-19 adjustment challenges’ positive association with GM only held true for Black participants (β = 0.160, p < .05) in the stratified model. Low distress tolerance’s positive association with GM held true across all racial groups in the stratified models: Latinx (β = −0.178, p < .05), Black (β = −0.211, p < .01), Asian (β = −0.138, p < .05), and White (β = −0.326, p < .001).

Table 5:

Predictors of gaming misuse (GM) stratified by race

Latinx n=372
Total adjusted R2 =.169		 Black n=368
Total adjusted R2 =.234		 Asian n=602
Total adjusted R2 =.187		 White n=439
Total adjusted R2 =.150
Unstandardized β Standardized β ΔR2 Unstandardized β Standardized β ΔR2 Unstandardized β Standardized β ΔR2 Unstandardized β Standardized β ΔR2
1: Sociodemographics 0.016 0.024 0.040* 0.014
Age 0.055 0.083 −0.044 −0.065 −0.003 −0.004 0.031 0.063
Gender (ref=men)
Women −0.632 −0.128 −0.795 −0.155* −1.003 −0.201*** −0.051 −0.014
Transgender or Other Gender 0.702 0.048 −0.263 −0.015 1.854 0.084 0.064 0.011
Sexual Orientation (ref=straight)
LGBA −0.34 −0.065 −0.506 −0.08 0.341 0.053 −0.026 −0.007
Questioning, prefer not to answer, or other 0.309 0.029 −1.491 −0.106 0.817 0.093 −0.326 −0.063
2: Pre-Existing Diagnoses 0.02 0.034* 0.014 0.014
Depression −0.343 −0.068 −0.057 −0.01 −0.553 −0.096 0.209 0.059
Anxiety 0.631 0.117 0.523 0.079 0.036 0.006 −0.433 −0.12
PTSD Diagnosis −0.671 −0.099 0.242 0.031 0.326 0.035 0.573 0.082
3: Psychological Symptoms 0.176*** 0.157*** 0.151*** 0.100**
Depression Symptoms 0.1 0.270* 0.023 0.062 0.147 0.378*** 0.047 0.153
Anxiety Symptoms −0.076 −0.208 0.015 0.039 −0.045 −0.107 −0.009 −0.026
PTSD Symptoms 0.049 0.370** 0.03 0.217 0.017 0.11 0.027 0.229
4: Distress 0.005 0.043** 0.008 0.036
COVID-19 Adjustment Challenges 0.003 0.008 0.059 0.160* 0.007 0.015 0.028 0.094
COVID-19 Stress 0.021 0.015 0.183 0.131 0.092 0.058 0.157 0.194
Financial Distress −0.143 −0.094 −0.13 −0.096 0.066 0.044 −0.118 −0.118
5: Distress Tolerance 0.017* 0.022* 0.010* 0.082***
Global Distress Tolerance −0.494 −0.178* −0.648 −0.211** −0.415 −0.138* −0.65 −0.326***
6: Social Connectedness 0.002 0.011 0.007 0
Social Support 0.03 0.017 −0.113 −0.07 0.114 0.052 −0.026 −0.017
Loneliness −0.062 −0.051 −0.175 −0.141 −0.102 −0.076 −0.019 −0.021
Open in a new tab
*

p < 0.05

**

p < 0.01

***

p < 0.001

4. Discussion

In our diverse sample of young adults during the COVID-19 pandemic, 11.5% of participants were classified as disordered social media users and 9.3% were classified as disordered gaming users as established by DSM-5 criteria of meeting at least five symptoms of Internet Gaming Disorder IGD. Due to limited existing research about PIU prevalence rates and varied definitions of PIU, comparing observed to existing rates is challenging. One review paper found a mean rate of “internet addiction” to be 8.2% in the U.S. and Europe (Weinstein & Lejoyeux, 2010). Our observed rate of disordered gamers is slightly higher than the 7.3% of problematic gamers found in a nationally representative sample in Norway (Wittek et al., 2016). Clarifying a consistent definition of PIU will be important for future research to determine reliable prevalence rates in a diverse population.

With respect to racial differences in rates, Latinx, Black, and Asian participants engaged in more SMM than White participants, and Black and Asian participants engaged in more GM than White participants. These differences are consistent with known lower rates of PIU among White adolescents and young adults compared to those of color (Liu et al., 2011; Nagata et al., 2022; Yates et al., 2012). Cultural differences may underlie these patterns, and future research is warranted to investigate this. Additionally, the internet offers safety and belonging to those with marginalized gender identities, and this effect may extend to those with racially minoritized identities, thereby resulting in increased internet use and possibly making racially minoritized individuals more vulnerable to excessive use (Austin et al., 2020).

4.1.0. Low Distress Tolerance Predicted PIU

Distress tolerance appeared to protect against PIU in our data; after controlling for sociodemographic predictors and mental health diagnoses and symptoms, we observed low distress tolerance to be most strongly and consistently associated with PIU. This finding also held true across most racial groups (with the exception of SMM for Latinx participants). Our findings are aligned with past research suggesting that distress tolerance is negatively associated with addictive behaviors and documents the relationship between distress tolerance and PIU during the uniquely stressful COVID-19 pandemic period. That those with low distress tolerance (i.e., those who felt distress more strongly) appeared to rely on PIU as a readily available coping tool during the pandemic is aligned with the stressful escapism model of PIU, in which overwhelming environmental stress is proposed to be associated with increased PIU (Snodgrass et al., 2014). Moreover, unlike other studies that focus on one form of PIU in relation to distress tolerance, we provide evidence of the association between low distress tolerance and two types of specific PIU – namely SMM and GM.

4.2.0. Psychological Distress and Associated Racial Differences in Predicting PIU

Notable findings emerged related to depression, PTSD, and COVID-19 adjustment challenges, in the overall and race stratified PIU models.

4.2.1. Depression.

Depression symptoms predicted increased GM, but not SMM. Race-stratified models showed a somewhat mixed pattern of associations. Depression symptoms predicted SMM and GM among Asian participants and only GM among Latinx participants. However, no association between depression symptoms and PIU was observed for White and Black participants. The links observed only for Asian and Latinx participants are consistent with prior findings which show that Asian and Latinx college students in the U.S. engage in more PIU than their Black and White peers (Liu et al., 2011). Our findings suggest that depression symptoms may drive different maladaptive behaviors in different racial groups, with depression symptoms appearing to drive PIU more in Asian and Latinx young adults than Black and White young adults in our sample. Although out of the scope of this study, depression symptoms among White and Black participants may drive different maladaptive coping behaviors. With existing research pointing to PIU as exacerbating future depression symptoms over time even while controlling for existing depression symptoms (Wartberg et al., 2018), this finding is concerning for Asian and Latinx participants. Nonetheless, more research is needed to understand why depression and PIU are not linked among other racial groups. We note that unlike symptoms, preexisting depression diagnosis was not associated with increased PIU.

4.2.2. PTSD.

In contrast to depression symptoms which were associated with GM in the overall sample, PTSD symptoms (while controlling for existing PTSD diagnoses) predicted increased SMM in the overall sample and across all racial groups. This is consistent with hypotheses and prior research about individual and societal-level traumatic experiences. In general, exposure to traumatic events increases one’s risk for developing addictive behaviors, including PIU (Hsieh et al., 2016). Specifically, more reported traumatic memories are linked to higher rates of PIU in an adolescent Italian sample, highlighting that PIU may assist in mental escapism from traumatic memories (Schimmenti et al., 2015). In terms of societal-level traumatic experiences and SMM, among survivors of Hurricane Sandy, increased posttraumatic stress was predictive of more social media use (Goodwin et al., 2013). Similarly, during a period of escalated violence in the West Bank of Palestine, posttraumatic symptoms were highest among those with the greatest frequency of social media use (Mahamid & Berte, 2018). Instead of escaping one’s traumatic memories through addictive behaviors such as PIU, adaptive coping with PTSD symptoms, such as creating a trauma narrative which assists memory consolidation (O’Kearney & Perrott, 2006), may prevent PIU in those with PTSD symptoms.

Contrary to our hypothesis, preexisting PTSD diagnosis predicted less SMM and was not associated with GM. In racially stratified models, this finding relating PTSD diagnoses to reduced SMM only held true among Latinx participants. There is no existing literature about differential PTSD experiences in Latinx individuals compared to others. While there may be something meaningful about having PTSD (e.g., avoidance of trauma triggers among Latinx individuals with PTSD), it is also possible that this finding reflects a statistical artifact. Future research is needed to understand the use of SMM and GM among those with a prior PTSD diagnosis.

4.2.3. COVID-19 Adjustment Challenges.

COVID-19 adjustment challenges also predicted more PIU among Black participants only and more SMM among Asian participants only. Using the stressful escapism model, this finding suggests that Black and Asian young adults experiencing COVID-19 adjustment challenges used PIU as a coping tool to combat overwhelming distress. Taken together, it appears that racial minoritized young adults are most vulnerable to coping with depression symptoms and adjustment-related distress with PIU compared to White young adults, who likely engage in alternative coping mechanisms in response to similar distress. Specifically, White young adults may be most likely to cope with depression though substance use, such as alcohol, tobacco, and prescription drug misuse, which are consistently elevated in White young adults compared to racially minoritized peers (Johnston et al., 2000; Herman-Stahl et al., 2006; Kroutil et al., 2006). Cultural norms related to coping behaviors may underlie these observed differences, and future research should investigate these reasons with methods that allow for in-depth explanation about reasons that contribute to behaviors (e.g., qualitative methods).

4.3.0. Notable demographic associations with PIU

Younger Latinx and White participants were more likely to engage in SMM, with no age-related differences observed in Black and Asian participants in SMM. PIU is more prevalent in younger individuals (Kuss et al., 2014; Nakayama et al., 2020). White women were more likely than White men to engage in SMM, which is consistent with a meta-analysis that found women to be more addicted to social media than men (Su et al., 2020). Men’s increased rates of GM compared to women only held true across Blacks and Asians; this is partially consistent with our hypothesis that all men would demonstrate higher rates of GM than women. While past work demonstrates that gaming rates are higher among males (Anderson et al., 2017; Kuss et al., 2014), future work should examine why gender differences in PIU vary across racial groups (e.g., perform intersectional research focused on race and gender to understand group differences). Sexual orientation did not predict PIU, with the exception that more Black lesbian, gay, bisexual, and asexual participants engage in more SMM than Black straight participants. With limited work focusing on SMM and sexual minority individuals, it is difficult to identify reasons underlying this difference. However, we do offer some speculation that the online world may be a place that fosters a sense of safety and belonging for those with marginalized or intersectional identities, a pattern that has been shown for gender minoritized individuals (Austin et al., 2020). While these experiences may be meaningful, they could also predispose individuals to greater SMM.

4.4.0. Implications

As PIU may be a newly classified mental disorder, it is timely to further develop an evidence-based understanding of the disorder’s etiology, risk factors, and clinical correlates such that (1) those at most risk can be identified and diagnosed, and (2) clinicians can intervene effectively to treat PIU. Our finding that low distress tolerance was associated with higher PIU provides support for the supposition that those with low distress tolerance should be closely monitored for PIU, and that fostering increased distress tolerance through evidence-based practices (i.e., mindfulness- and exposure-based interventions; Lotan et al., 2013; Nila et al., 2016) should be a clinical priority. Findings that PTSD symptoms predict SMM, and depression symptoms predict GM highlight the importance of treating present PTSD and/or depression symptoms in patients with SMM and/or GM. For example, if a patient with a recent traumatic experience presenting with PTSD symptoms is engaging in excessive social media use, clinicians should focus on evidence-based practices to address PTSD symptoms, such as addressing avoidance of traumatic memories with trauma narrative work (O’Kearney & Perrott, 2006). This in turn may decrease SMM. Our findings also highlight the importance of screening for PIU in those with depression and PTSD symptoms. Finally, our findings underscore the importance of ongoing symptom assessment. That preexisting diagnoses are less robust predictors of PIU than current symptoms call clinicians to engage in regular symptom checklists with patients, particularly during times of collective distress.

4.4.0. Strengths and Limitations

A strength of this study is its sample, which was large and racially diverse; this allowed for prevalence rate comparisons between racial groups instead of White versus non-White in the overall model. This also allowed for stratification by race, which provided a nuanced understanding of how predictors of PIU varied or remained constant between four racial groups. Measurement of mental health diagnoses and current symptoms allowed for control of preexisting diagnoses and examination of how current psychological distress related to PIU. By focusing on common COVID-19 stressors – financial distress, loneliness, social support, and stress and adjustment challenges directly related to COVID-19 – we were able to examine the specific type of distress associated with PIU during the pandemic.

While a strength of our study is its context-specific variables that relate to the experiences of participants during the COVID-19 pandemic, this may also limit prospective generalizability of findings to other time contexts. Given the interest in Internet Gaming Disorder for consideration in a future version of DSM, future research should examine how distress following the pandemic relates to PIU. As our cross-sectional analyses did not allow for identification of directionality between variables included in our models, future studies should use longitudinal methods to determine temporal relationships between PIU behaviors and variables such as PTSD and depression symptoms. Uncovering if PIU is primarily a response to or a driver of psychological symptoms will provide clinically important information for treating PIU and other mental disorders. Moreover, self-report of PIU was also likely impacted by biases inherent to self-report measures. To address this limitation, future research that utilizes smartphone tracking data may increase validity related to social media use.

Additionally, although our measure of GM aligns with the proposed IGD criteria, we adapted it from a validated measure of SMM. Thus, although reliability of the adapted measure in our sample was good, no additional psychometric properties, such as validity or additional measures of reliability, are known. Future research is warranted to validate this adapted measure of GM and its associated clinical cut-off point. In addition, while we based the variables we included in analyses on existing research and theory, it is also possible that we excluded variables that further helped explain PIU in our analyses. Investment in future ongoing quantitative and qualitative research related to PIU is critical to ensure that all variables that relate to PIU are identified. While our sample size was robust insofar as it allowed for stratified analyses across four racial groups, we were unable to include smaller racial groups in the U.S. in these exploratory analyses, such as Native Americans, who may display unique patterns related to PIU. To build a comprehensive understanding of how PIU may vary by race, inclusion of all racial groups is warranted in future research. As our survey was only offered in English, our sample was also limited to English speakers, which may reduce the external validity of our findings.

5. Conclusion

With increasing access to the internet and smartphones among young adults in the U.S., PIU is considered a public health problem (Block, 2008) associated with significant consequences in terms of mental health, everyday functioning, and quality of life (Lozano-Blasco et al., 2022; Spanda, 2014). Nonetheless, definitions of PIU vary greatly in the literature, PIU is not yet a diagnosable mental disorder, and research about the etiology of the disorder and associated risk factors has notable gaps. During the COVID-19 time of unprecedented environmental stress and increased digitalization (Amankwah-Amoah et al., 2021), we investigated COVID-19-related and other empirically derived risk factors for PIU in a young adult sample to assess which factors most contributed to PIU.

Depression, PTSD symptoms, COVID-19 stress and adjustment challenges, and low distress tolerance experienced during the pandemic were associated with increased PIU. Taken together, it appears that environmental stressors around COVID-19 and clinical symptoms drove PIU during the pandemic in young adults, with those less able to tolerate distress engaging in more PIU. This suggests that the stressful escapism model of PIU is a relevant model through which to understand PIU’s etiology during the pandemic period of collective distress. Future research using larger samples should examine methods of escapism from stress more broadly to further delineate factors (e.g., race, mental health) that predict different types of maladaptive coping. This would provide valuable data to researchers, clinicians, and those considering the addition of Internet Gaming Disorder as a diagnostic category in DSM.

Acknowledgements and Funding Disclosure

GAW conceived the initial idea, designed the study, analyzed and interpreted the findings, conducted the literature review, and wrote the manuscript. CS contributed to study design, interpretation of findings, relevant literature review, and reviewing multiple drafts of the manuscript. HCH contributed to study design, interpretation of findings, and reviewing multiple drafts of the manuscript. CHL assisted with conception of the initial idea, study design, interpretation of findings, analyses, and reviewing multiple drafts of the manuscript. This research was funded by the National Science Foundation (#2027553) awarded to CHL and HCH and a training fellowship for GAW through the National Institutes of Health (T32 MH016259).

Footnotes

Disclosures and Conflicts of Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References

  1. Aboujaoude E (2010). Problematic Internet use: an overview. World Psychiatry, 9(2), 85. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Alheneidi H, AlSumait L, AlSumait D, & Smith AP (2021). Loneliness and problematic internet use during COVID-19 lock-down. Behavioral Sciences, 11(1), 5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Amankwah-Amoah J, Khan Z, Wood G, & Knight G (2021). COVID-19 and digitalization: The great acceleration. Journal of Business Research, 136, 602–611. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). [Google Scholar]
  5. Anderson EL, Steen E, & Stavropoulos V (2017). Internet use and problematic internet use: A systematic review of longitudinal research trends in adolescence and emergent adulthood. International Journal of Adolescence and Youth, 22(4), 430–454. [Google Scholar]
  6. Andrykowski MA, Cordova MJ, Studts JL, & Miller TW (1998). Posttraumatic stress disorder after treatment for breast cancer: Prevalence of diagnosis and use of the PTSD Checklist—Civilian Version (PCL-C) as a screening instrument. Journal of Consulting and Clinical Psychology, 66(3), 586. [DOI] [PubMed] [Google Scholar]
  7. Austin A, Craig SL, Navega N, & McInroy LB (2020). It’s my safe space: The life-saving role of the internet in the lives of transgender and gender diverse youth. International Journal of Transgender Health, 21(1), 33–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bányai F, Zsila Á, Király O, Maraz A, Elekes Z, Griffiths MD, ... & Demetrovics Z (2017). Problematic social media use: Results from a large-scale nationally representative adolescent sample. PloS One, 12(1), e0169839. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Block JJ (2008). Issues for DSM-V: Internet addiction. American Journal of Psychiatry, 165(3), 306–307. [DOI] [PubMed] [Google Scholar]
  10. Bornovalova MA, Gratz KL, Daughters SB, Hunt ED, & Lejuez CW (2012). Initial RCT of a distress tolerance treatment for individuals with substance use disorders. Drug and Alcohol Dependence, 122(1–2), 70–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Broderick PC, & Jennings PA (2012). Mindfulness for adolescents: A promising approach to supporting emotion regulation and preventing risky behavior. New Directions for Youth Development, 2012(136), 111–126. [DOI] [PubMed] [Google Scholar]
  12. Caplan SE (2002). Problematic Internet use and psychosocial well-being: development of a theory-based cognitive–behavioral measurement instrument. Computers in Human Behavior, 18(5), 553–575. [Google Scholar]
  13. Caplan SE (2006). Relations among loneliness, social anxiety, and problematic Internet use. CyberPsychology & Behavior, 10(2), 234–242. [DOI] [PubMed] [Google Scholar]
  14. Chae DH, Yip T, Martz CD, Chung K, Richeson JA, Hajat A, ... & LaVeist TA (2021). Vicarious racism and vigilance during the COVID-19 pandemic: mental health implications among Asian and Black Americans. Public Health Reports, 136(4), 508–517. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Charlton PFC, & Thompson JA (1996). Ways of coping with psychological distress after trauma. British Journal of Clinical Psychology, 35(4), 517–530. [DOI] [PubMed] [Google Scholar]
  16. Demetrovics Z, Király O, Koronczai B, Griffiths MD, Nagygyörgy K, Elekes Z, ... & Urbán R (2016). Psychometric properties of the Problematic Internet Use Questionnaire Short-Form (PIUQ-SF-6) in a nationally representative sample of adolescents. PloS one, 11(8), e0159409. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Dhira TA, Rahman MA, Sarker AR, & Mehareen J (2021). Validity and reliability of the Generalized Anxiety Disorder-7 (GAD-7) among university students of Bangladesh. PloS One, 16(12), e0261590. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Dillman Carpentier FR, Brown JD, Bertocci M, Silk JS, Forbes EE, & Dahl RE (2008). Sad kids, sad media? Applying mood management theory to depressed adolescents’ use of media. Media Psychology, 11(1), 143–166. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Elhai JD, Levine JC, O’Brien KD, & Armour C (2018). Distress tolerance and mindfulness mediate relations between depression and anxiety sensitivity with problematic smartphone use. Computers in Human Behavior, 84, 477–484. [Google Scholar]
  20. Fernandes B, Biswas UN, Mansukhani RT, Casarín AV, & Essau CA (2020). The impact of COVID-19 lockdown on internet use and escapism in adolescents. Revista de psicología clínica con niños y adolescentes, 7(3), 59–65. [Google Scholar]
  21. Goodwin R, Palgi Y, Hamama-Raz Y, & Ben-Ezra M (2013). In the eye of the storm or the bullseye of the media: social media use during Hurricane Sandy as a predictor of posttraumatic stress. Journal of Psychiatric Research, 47(8), 1099–1100. [DOI] [PubMed] [Google Scholar]
  22. Grey I, Arora T, Thomas J, Saneh A, Tohme P, & Abi-Habib R (2020). The role of perceived social support on depression and sleep during the COVID-19 pandemic. Psychiatry Research, 293, 113452. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Horwitz AG, Hill RM, & King CA (2011). Specific coping behaviors in relation to adolescent depression and suicidal ideation. Journal of Adolescence, 34(5), 1077–1085. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Hosseinian S, & Nooripour R (2019). Effectiveness of mindfulness-based intervention on risky behaviors, resilience, and distress tolerance in adolescents. International Journal of High Risk Behaviors and Addiction, 8(4). [Google Scholar]
  25. Hsieh YP, Shen ACT, Wei HS, Feng JY, Huang SCY, & Hwa HL (2016). Associations between child maltreatment, PTSD, and internet addiction among Taiwanese students. Computers in Human Behavior, 56, 209–214. [Google Scholar]
  26. Hughes ME, Waite LJ, Hawkley LC, & Cacioppo JT (2004). A short scale for measuring loneliness in large surveys: Results from two population-based studies. Research on Aging, 26(6), 655–672. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Jarman HK, Marques MD, McLean SA, Slater A, & Paxton SJ (2021). Motivations for social media use: Associations with social media engagement and body satisfaction and well-being among adolescents. Journal of Youth and Adolescence, 50(12), 2279–2293. [DOI] [PubMed] [Google Scholar]
  28. Jouhki H, Savolainen I, Sirola A, & Oksanen A (2022). Escapism and excessive online behaviors: a three-wave longitudinal study in Finland during the COVID-19 pandemic. International Journal of Environmental Research and Public Health, 19(19), 12491. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Kozak AT, & Fought A (2011). Beyond alcohol and drug addiction. Does the negative trait of low distress tolerance have an association with overeating?. Appetite, 57(3), 578–581. [DOI] [PubMed] [Google Scholar]
  30. Kraut R, Lundmark V, Patterson M, Kiesler S, Mukopadhyahy T, & Sherlis W, 1998. Internet paradox: A social technology that reduces social involvement and psychological well being? American Psychologist, 53, 1017–1031. [DOI] [PubMed] [Google Scholar]
  31. Kuss DJ, Griffiths MD, Karila L, & Billieux J (2014). Internet addiction: A systematic review of epidemiological research for the last decade. Current Pharmaceutical Design, 20, 4026–4052. [DOI] [PubMed] [Google Scholar]
  32. Laconi S, Tricard N, & Chabrol H (2015). Differences between specific and generalized problematic Internet uses according to gender, age, time spent online and psychopathological symptoms. Computers in Human Behavior, 48, 236–244. [Google Scholar]
  33. Lazarus RS, & Folkman S (1984). Stress, Appraisal, and Coping. Springer publishing company. [Google Scholar]
  34. Lee BW, & Stapinski LA (2012). Seeking safety on the internet: Relationship between social anxiety and problematic internet use. Journal of Anxiety Disorders, 26(1), 197–205. [DOI] [PubMed] [Google Scholar]
  35. Liu TC, Desai RA, Krishnan-Sarin S, Cavallo DA, & Potenza MN (2011). Problematic Internet use and health in adolescents: data from a high school survey in Connecticut. The Journal of Clinical Psychiatry, 72(6), 15663. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Lopez-Fernandez O (2018). Generalised versus specific internet use-related addiction problems: A mixed methods study on internet, gaming, and social networking behaviours. International Journal of Environmental Research and Public Health, 15(12), 2913. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Lotan G, Tanay G, & Bernstein A (2013). Mindfulness and distress tolerance: relations in a mindfulness preventive intervention. International Journal of Cognitive Therapy, 6(4), 371–385. [Google Scholar]
  38. Lozano-Blasco R, Robres AQ, & Sánchez AS (2022). Internet addiction in young adults: A meta-analysis and systematic review. Computers in Human Behavior, 130, 107201. [Google Scholar]
  39. Lukavská K, Hrabec O, Lukavský J, Demetrovics Z, & Király O (2022). The associations of adolescent problematic internet use with parenting: A meta-analysis. Addictive Behaviors, 107423. [DOI] [PubMed] [Google Scholar]
  40. Lund S, Ellingrud K, Hancock B, & Manyika J (2020). COVID-19 and jobs: Monitoring the US impact on people and places. McKinsey Global Institute. [Google Scholar]
  41. Mahamid FA, & Berte DZ (2020). Portrayals of violence and at-risk populations: Symptoms of trauma in adolescents with high utilization of social media. International Journal of Mental Health and Addiction, 18, 980–992. [Google Scholar]
  42. Masaeli N, & Farhadi H (2021). Prevalence of Internet-based addictive behaviors during COVID-19 pandemic: A systematic review. Journal of Addictive Diseases, 39(4), 468–488. [DOI] [PubMed] [Google Scholar]
  43. Mihara S, Osaki Y, Nakayama H, Sakuma H, Ikeda M, Itani O, ... & Higuchi S (2016). Internet use and problematic Internet use among adolescents in Japan: A nationwide representative survey. Addictive Behaviors Reports, 4, 58–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Mitchell BJ, Gawlik EA, Baugher BJ, George RL, Muakkassa FF, Mallat AF, ... & Coifman KG (2022). Were there losses in social support during the pandemic? Testing the impact of COVID-19 on psychological adjustment to trauma in United States adults. Frontiers in Psychology, 13, 1061621. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Moretta T, & Buodo G (2020). Problematic Internet use and loneliness: How complex is the relationship? A short literature review. Current Addiction Reports, 7, 125–136. [Google Scholar]
  46. Nagata JM, Singh G, Sajjad OM, Ganson KT, Testa A, Jackson DB, ... & Baker FC (2022). Social epidemiology of early adolescent problematic screen use in the United States. Pediatric Research, 92(5), 1443–1449. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Nakayama H, Ueno F, Mihara S, Kitayuguchi T, & Higuchi S (2020). Relationship between problematic Internet use and age at initial weekly Internet use. Journal of Behavioral Addictions, 9(1), 129–139. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Nila K, Holt DV, Ditzen B, & Aguilar-Raab C (2016). Mindfulness-based stress reduction (MBSR) enhances distress tolerance and resilience through changes in mindfulness. Mental Health & Prevention, 4(1), 36–41. [Google Scholar]
  49. O’Kearney R, & Perrott K (2006). Trauma narratives in posttraumatic stress disorder: A review. Journal of traumatic stress, 19(1), 81–93. [DOI] [PubMed] [Google Scholar]
  50. Otto MW, Powers MB, & Fischmann D (2005). Emotional exposure in the treatment of substance use disorders: Conceptual model, evidence, and future directions. Clinical Psychology Review, 25(6), 824–839. [DOI] [PubMed] [Google Scholar]
  51. Otto MW, Safren SA, & Pollack MH (2004). Internal cue exposure and the treatment of substance use disorders: Lessons from the treatment of panic disorder. Journal of Anxiety Disorders, 18(1), 69–87. [DOI] [PubMed] [Google Scholar]
  52. Park CL, Russell BS, Fendrich M, Finkelstein-Fox L, Hutchison M, & Becker J (2020). Americans’ COVID-19 stress, coping, and adherence to CDC guidelines. Journal of General Internal Medicine, 35, 2296–2303. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Pew Research Center. (2022). Teens, social media, and technology 2022. [Google Scholar]
  54. Pressler SJ, Subramanian U, Perkins SM, Gradus-Pizlo I, Kareken D, Kim J, ... & Sloan R (2011). Measuring depressive symptoms in heart failure: Validity and reliability of the Patient Health Questionnaire–8. American Journal of Critical Care, 20(2), 146–152. [DOI] [PubMed] [Google Scholar]
  55. Refaeli T, & Achdut N (2021). Financial strain and loneliness among young adults during the COVID-19 pandemic: The role of psychosocial resources. Sustainability, 13(12), 6942. [Google Scholar]
  56. Schimmenti A, Passanisi A, Caretti V, La Marca L, Granieri A, Iacolino C, ... & Billieux J (2017). Traumatic experiences, alexithymia, and Internet addiction symptoms among late adolescents: A moderated mediation analysis. Addictive Behaviors, 64, 314–320. [DOI] [PubMed] [Google Scholar]
  57. Shapira NA, Goldsmith TD, Keck PE Jr, Khosla UM, & McElroy SL (2000). Psychiatric features of individuals with problematic internet use. Journal of Affective Disorders, 57(1–3), 267–272. [DOI] [PubMed] [Google Scholar]
  58. Skues J, Williams B, Oldmeadow J, & Wise L (2016). The effects of boredom, loneliness, and distress tolerance on problem internet use among university students. International Journal of Mental Health and Addiction, 14, 167–180. [Google Scholar]
  59. Simons JS, & Gaher RM (2005). The Distress Tolerance Scale: Development and validation of a self-report measure. Motivation and Emotion, 29, 83–102. [Google Scholar]
  60. Singh S, Datta M, Gupta P, & Batra S (2022). Predictors of ‘problematic internet use’ among adolescents and adults amid the pandemic in India. Clinical Epidemiology and Global Health, 15, 101036. [Google Scholar]
  61. Snodgrass JG, Lacy MG, Dengah II HF, Eisenhauer S, Batchelder G, & Cookson RJ (2014). A vacation from your mind: Problematic online gaming is a stress response. Computers in Human Behavior, 38, 248–260. [Google Scholar]
  62. Son M, & Jeong GC (2023). Influence of Sensation Seeking and Life Satisfaction Expectancy on Stock Addiction Tendency: Moderating Effect of Distress Tolerance. Behavioral Sciences, 13(5), 378. [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Spada MM (2014). An overview of problematic Internet use. Addictive Behaviors, 39(1), 3–6. [DOI] [PubMed] [Google Scholar]
  64. Spitzer RL, Kroenke K, Williams JB, & Löwe B (2006). A brief measure for assessing generalized anxiety disorder: the GAD-7. Archives of Internal Medicine, 166(10), 1092–1097. [DOI] [PubMed] [Google Scholar]
  65. Stenseng F, Falch-Madsen J, & Hygen BW (2021). Are there two types of escapism? Exploring a dualistic model of escapism in digital gaming and online streaming. Psychology of Popular Media, 10(3), 319. [Google Scholar]
  66. Stevens C, Zhang E, Cherkerzian S, Chen JA, & Liu CH (2020). Problematic internet use/computer gaming among US college students: Prevalence and correlates with mental health symptoms. Depression and Anxiety, 37(11), 1127–1136. [DOI] [PMC free article] [PubMed] [Google Scholar]
  67. Su W, Han X, Yu H, Wu Y, & Potenza MN (2020). Do men become addicted to internet gaming and women to social media? A meta-analysis examining gender-related differences in specific internet addiction. Computers in Human Behavior, 113, 106480. [Google Scholar]
  68. Tham SM, Ellithorpe ME, & Meshi D (2020). Real-world social support but not in-game social support is related to reduced depression and anxiety associated with problematic gaming. Addictive Behaviors, 106, 106377. [DOI] [PubMed] [Google Scholar]
  69. Van den Eijnden RJ, Lemmens JS, & Valkenburg PM (2016). The social media disorder scale. Computers in Human Behavior, 61, 478–487. [Google Scholar]
  70. Wartberg L, Kriston L, Zieglmeier M, Lincoln T, & Kammerl R (2018). A longitudinal study on psychosocial causes and consequences of internet gaming disorder in adolescence. Psychological Medicine, 49, 287–294. [DOI] [PubMed] [Google Scholar]
  71. Weinstein A, & Lejoyeux M (2010). Internet addiction or excessive internet use. The American Journal of Drug and Alcohol Abuse, 36(5), 277–283. [DOI] [PubMed] [Google Scholar]
  72. Wittek CT, Finserås TR, Pallesen S, Mentzoni RA, Hanss D, Griffiths MD, & Molde H (2016). Prevalence and predictors of video game addiction: A study based on a national representative sample of gamers. International Journal of Mental Health and Addiction, 14, 672–686. [DOI] [PMC free article] [PubMed] [Google Scholar]
  73. Xu S, Park M, Kang UG, Choi JS, & Koo JW (2021). Problematic use of alcohol and online gaming as coping strategies during the COVID-19 pandemic: a mini review. Frontiers in Psychiatry, 12, 685964. [DOI] [PMC free article] [PubMed] [Google Scholar]
  74. Yan W, Li Y, & Sui N (2014). The relationship between recent stressful life events, personality traits, perceived family functioning and internet addiction among college students. Stress and Health, 30(1), 3–11. [DOI] [PubMed] [Google Scholar]
  75. Yates TM, Gregor MA, & Haviland MG (2012). Child maltreatment, alexithymia, and problematic internet use in young adulthood. Cyberpsychology, Behavior, and Social Networking, 15, 219–225. doi: 10.1089/cyber.2011.0427 [DOI] [PubMed] [Google Scholar]
  76. Zhao N, & Zhou G (2021). COVID-19 stress and addictive social media use (SMU): Mediating role of active use and social media flow. Frontiers in Psychiatry, 12, 635546. [DOI] [PMC free article] [PubMed] [Google Scholar]
  77. Zimet GD, Dahlem NW, Zimet SG, & Farley GK (1988). The multidimensional scale of perceived social support. Journal of Personality Assessment, 52(1), 30–41. [DOI] [PubMed] [Google Scholar]

RESOURCES