COVID-19 pandemic impacted internet use and anxiety among general public during COVID-19 pandemic in Pakistan

Objective: To determine the association between internet use and anxiety among people during the coronavirus disease-2019 pandemic. Method: The cross-sectional study was conducted across Pakistan from January 14 to February 21, 2021, which was the active phase of the coronavirus disease-2019 pandemic in Pakistan. The participants were aged at least 13 years having internet access regardless of gender or their location across Pakistan. The anonymous web-based survey was conducted using a questionnaire generated on Google Forms and disseminated through various social media platforms and WhatsApp groups. Anxiety symptoms were screened using the Depression, Anxiety and Stress Scale-21, while the Young Internet Addiction Test was used to evaluate symptoms of internet addiction. Data was analysed using STATA 16. Results: Of the 1,145 subjects, 686(60%) were females and 459(40%) were males. A total of 257(22.5%) participants were found to have extremely severe anxiety and internet usage pattern was significantly associated with the level of anxiety ( p <0.05). Age, gender, social class and marital status were not significantly diﬀerent ( p >0.05), while family income and area of living were significantly diﬀerent ( p <0.05) in terms of anxiety levels. The odd of addictive internet use was 10.2 (95% confidence interval: 5.7-18.5) times greater in extreme anxiety individuals compared to individuals having no anxiety after controlling for other sociodemographic, health-related, behavioural and environmental factors during the pandemic. Conclusion: A significant association of anxiety was found with internet addiction during the coronavirus disease-2019 pandemic.


Introduction
The coronavirus disease-2019 (COVID- 19), undoubtedly the greatest public health catastrophe of modern times and a universal health concern in its active phase, brought a number of challenges in all domains of life, including the healthcare systems. 1 Quarantine and social distancing further intensified the misery, subsequently causing negative emotions in people. 2 Moreover, the public health measures undertaken to contain the spread of the lethal virus, including massive lockdowns, also negatively impacted people's quality of life. 3COVID-19 was not just a medical phenomenon affecting people physically.It affected the mental wellbeing of people as well, causing a massive psychological strain. 4This could actually mean that in a world where mental health problems were already on the rise, the pandemic could have proven to be a catalyst.Meta-analysis of 103 studies across the world indicated alarming levels of symptoms of mental illness, predominantly anxiety, during COVID-19, with a prevalence of 27.3% in the uninfected general population, and 39.6% in the infected population. 1 Subgroup analysis of a systematic review suggested the burden of anxiety during COVID-19 ranging from 28.3% to 54.6% in the developed world. 1 In South Asia, the pooled prevalence of anxiety during the pandemic was found to be 34.1%. 5In India, the burden of anxiety symptoms was 38.2% during the lockdown phase 6 whereas in Bangladesh, the prevalence of anxiety was 57.9%. 7There is very limited literature on the prevalence of anxiety symptoms in Pakistan during the pandemic in the general population, with most studies havingf catered to the mental health of healthcare workers or the students. 8,9However, one study conducted on the general population suggested an alarming prevalence of 40% of moderate and severe anxiety symptoms in the general population. 10xhaustion, attention deficit and agitation. 11During the pandemic, there were multitudes of factors related to anxiety.One of the most contributing factors associated with anxiety was excessive use of the internet or internet addiction. 12,13Internet addiction is characterised by excessive use of the internet, the urge to frequently use the internet, repeated efforts to stop using the internet, loss of importance of time using the internet, and extreme uneasiness when not being able to use the internet. 14uring COVID-19, information transmission through digital means-conventionally called the internet-became the only accessible method for the total non-interruption of social and work collaborations in an attempt to restore a new form of normality.The excessive use and high dependency on the internet might be the reason for internet addiction during the pandemic.Multiple studies across the world have indicated an alarming burden of internet addiction during COVID-19.6][17] Statistics from India suggest 20% moderate to severe internet addiction. 18ring the pandemic, internet usage rose, and anxiety was likely to creep in.A study showed that in India, Malaysia, Mexico and the United Kingdom, participants scoring high on compulsive internet and social media use had much higher anxiety scores. 19Another study in Mexico indicated that participants with internet addiction were twice likely to have anxiety. 12Literature on Italian students also suggests an association between internet addiction and anxiety mediated by the fear of COVID-19. 20cessive use of the internet has been associated with the spread of misleading information, causing fear, stress and worries that might ultimately result in anxiety and impaired functioning.Also, to verify the misinformation, internet usage also increases, which leads to a vicious cycle of internet addiction and mental health issues. 12,21kistan is among the lower-and middle income countries (LMICs) with finite resources.Preventive and curative treatment of mental illnesses is not a priority in the Pakistani health setup.The advent of COVID-19 shook up the already exhausted mental health infrastructure in the country.To prevent the conversion of initial psychological symptoms to full-blown mental disorders, it is essential that their burden and associated factors are identified and proactively tackled.The current study was planned to determine the burden of anxiety and explore its associated factors in the Pakistani population during the COVID-19 pandemic.

Subjects and Methods
The online web-based cross-sectional study was conducted from January 14 to February 21, 2021, which was the active phase of the pandemic in Pakistan.After approval from the institutional ethics review committee, the sample size was calculated using OpenEpi calculator, with 80% power and 5% level of significance while assuming the prevalence of psychological morbidity in normal internet users to be 28-33%. 22The odds ratio was anticipated to be 1.5.The sample was inflated by 15% to account for refusals and incomplete forms.Additional details on the sample size can be found in the 'Sampling Size and Sampling Strategy' section of an earlier published article. 23ose included were native Pakistanis of either gener aged at least 13 years who were familiar with Urdu or English languages, and were willing to participate.Overseas Pakistanis and foreign nationals living in Pakistan were excluded.
Informed consent was obtained from all the particpants.In cases of participants aged <18 years, parental consent was taken, and contact details of parents/guardians were taken to verify the consent.The consent form and the study questionnaire were in both English and Urdu languages.To ensure cultural and contextual relevance, content validity was done.A psychiatrist reviewed and translated the English version of the survey into Urdu; The translated Urdu version was then back-translated into English to make sure answers were relevant and comparable in meaning and context.
The study questionnaire was generated on Google Forms, and a relevant link was generated to allow access to that form.The link was disseminated via social media platforms, including Facebook and Instagram, as well through WhatsApp and emails.
The dependent variable in the study was anxiety symptoms that were screened using the Depression, Anxiety and Stress Scale-21 (DASS-21), which employs the symptomatology approach to screen for psychological symptoms, but is not used for a thorough diagnosis. 22The anxiety subscale of DASS-21 has superlative psychometric properties, with Cronbach alpha 0.89 indicating excellent internal consistency, convergent and concurrent validities. 24The scale has been previously used and validated in the Pakistani population. 25l items were scored on a 4-point Likert scale, , with 0=it does not apply to me at all, 1=it applies to me to some degree or some of the time, 2=it applies to me to a considerable degree or a good part of the time, and 3=it applies to me very much, or most of the time.Summation  The 20 IAT items were scored on a 6-point Likert scale, with 0=not applicable, 1=rarely, 2=occasionally, 3=frequently, 4=often and 5=always.The total score ranged 0-100, with 0-39=normal internet users (NIUs), 40-69=roblematic internet users (PIUs), and 70 and above=addictive internet users (AIUs).
Other independent variables included sociodemographic factors, family and medical history, and behavioural and environmental factors related to COVID-19.
Data was analysed using STATA 16.Mean and standard deviations were reported for continuous quantitative variables, and frequencies and percentages were reported for categorical variables.Multinomial regression was used to determine associated factors of anxiety, and odds ratios (ORs) along with 95% confidence intervals (CIs) were reported.P<0.05 was considered statistically significant.

Results
Of the 1,145 subjects, 686(60%) were females and 459(40%) were males.There were 454(39.7%)subjects with no anxiety, 109(9.5%)had mild anxiety, 242(21.1%)had moderate, 83(7.2%) had severe, and 257(22.5%)had extremely severe anxiety.Internet usage pattern among NIUs, PIUs and AIUs was significantly associated with the level of anxiety (p<0.05).Age, gender, social class and marital status were not significantly different (p>0.05),while family income and area of living were significantly different (p<0.05) in terms of anxiety levels.Behavioural, environmental and health-related factors were also significantly different (Table 1).The odd of addictive internet use was 10.2 (95% confidence interval: 5.7-18.5)times greater in individuals with extreme anxiety compared to individuals having no anxiety after controlling for other sociodemographic, health-related, behavioural and environmental factors during the pandemic (Table 2).

Discussion
About 30% of the participants in the current study showed severe to extremely severe anxiety levels, indicating an increase in anxiety levels among Pakistani population during the COVID-19 pandemic.This coincides with several studies done in China, Turkey and Mexico. 12,28-30A systematic review and meta-analysis of 17 studies also revealed that anxiety levels during the pandemic had increased to about 32%. 31 Internet addiction was found to be one of the significant factors associated with anxiety levels in the present study, which is consistent with literature. 32,33Moreover, lockdowns during the pandemic contributed markedly in increased usage of internet 34,35 which could possibly explain high anxiety levels in the population in general. 29e current study also found that females experienced higher levels of anxiety than males.This finding was supported by a study in Saudi Arabia. 36other important finding of the current study was that people suffering from mental problems or chronic illness had greater anxiety levels.People with pre-existing mental health problems or with any other chronic illness found it especially difficult to deal with the pandemic.Evidence from a meta-analysis demonstrated an increased risk of fatal COVID-19 among patients with a pre-existing mental disorder 37 and this fear could be the possible explanation of higher anxiety levels among them.
A major strength of the current study is that it was conducted during the peak phase of COVID-19.
Additionally, an adequate and large sample size was obtained due to the usage of validated tools and easyaccess survey distributed through social media. 102 Open Access J Pak Med Assoc The current study has some limitations.As it was an internet-based survey, the sample, therefore, may not be representative of the entire population.Moreover, data was collected using a self-administered tool, and the anxiety levels may not be compatible with the formal confirmatory diagnosis by a mental health professional.Also, the crosssectional survey has a potential inherit bias as the outcome was assessed at one point in time.Any temporal or causal association cannot be established, and a time trend analysis is recommended for future studies.

Conclusion
A significant association of anxiety was found with internet addiction during the COVID-19 pandemic.Relevant strategies, including counselling, should be curated to curb internet addiction as it can lead to mental health issues that might culminate into mental disorders.

Table - 1
: Baseline characteristics of the participants as per the degree of anxiety symptoms.

Table - 1
: Continued from previous page.

Table - 2
: Predictors of different levels of anxiety..