Internet addiction (IA) is characterized by inability to control Internet use which interferes with one's daily life.¹ IA is not listed in the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5). However, Internet Gaming Disorder is listed in Section III as a condition warranting more scientific research and clinical experience before it might be considered as a formal disorder.² IA broadly includes other types of addictive internet use (i.e., addictive use of social networking sites, addictive online shopping, and excessive internet information retrieval) and internet gaming.³

The prevalence of IA varies among populations. It ranges from 0.3%⁴ to 37%.⁵ This variability across studies might have resulted from variations in Internet accessibility in different countries, definitions of IA, assessment tools, and age ranges of different population.⁶ Significant association between IA and psychiatric comorbidity has been documented in numerous studies (for a review³⁵⁷). However, most studies have been focused on groups of students or adolescents in school-based studies or Internet users in online surveys. Ho et al.³ have conducted a meta-analysis of the literature and determined the association between IA and psychiatric comorbidity including 8 studies comprising 1,641 patients with IA and 11,210 controls. Positive associations between IA and alcohol abuse, attention-deficit hyperactivity disorder, depression, and anxiety have been found, with pooled prevalence rates ranging from 13.5% for alcohol abuse to 26.3% for depression.³ To the best of our knowledge, there are only two published studies on the prevalence and comorbidities of IA in a random sample of the general population. Aboujaoude et al.⁴ have reported a prevalence of problematic Internet use of 0.3% in a general population of the United States (≥18 years of age, n=2,513). Bakken et al.⁸ have reported a prevalence of IA of 1.0% in the general Norwegian population (16-74 years, n=3,399) with a linear relationship between the prevalence of self-reported mental health problems and increasing Internet-related problems.

Given the paucity of using general population of random adults sample to study IA, we investigated the prevalence and psychiatric comorbidities of IA in a nationally representative sample of adults aged 18-70 years in South Korea.

Of the total of 3,723 participants, 39 (1.0%, 95% CI: 0.7-1.3%) were classified as addicted Internet users. Regarding the main area of Internet use, addictive Internet users were more likely to use social media (20.5% vs. 8.8%, χ²=4.89, p=0.027) and Internet game (33.3% vs. 12.2%, χ²=15.81, p<0.001) compared to usual Internet users. They were less likely to use Internet for information retrieval (34.2% vs. 56.2%, χ²=7.41, p=0.006) compared to usual Internet users.

Addictive Internet users were younger (p=0.009) than usual Internet users. They were more likely to be male (p=0.027) and unemployed (p=0.006) compared to usual Internet users. Educational status and economic status were not significantly different between the two groups. A total of 74.4% of addicted Internet users and 29.1% of usual Internet users had at least one DSM-IV psychiatric disorder (Table 1).

IA was positively associated with a wide range of psychiatric comorbidities, including alcohol [adjusted odd ratio (AOR)= 2.10, 95% CI=1.02-4.29, p=0.043] and nicotine use disorders (AOR=3.98, 95% CI=1.82-8.69, p=0.001), depressive disorders (AOR=9.67, 95% CI=4.86-19.23, p<0.001), anxiety disorder (AOR=8.17, 95% CI=4.14-16.09, p<0.001), and any DSM-IV psychiatric disorder (AOR=6.94, 95% CI=3.32-14.55, p<0.001) after controlling for age, gender, and occupational status. When anxiety disorder was classified into several more specific types, IA was significantly and positively associated with generalized anxiety disorder (AOR=15.39, 95% CI=6.16-38.43, p<0.001), specific phobia (AOR=2.76, 95% CI=1.03-7.36, p=0.043), social phobia (AOR=7.02, 95% CI=1.34-36.88, p=0.021), and post-traumatic stress disorder (AOR=8.32, 95% CI=2.37-29.25, p=0.001).

The present study provided the first representative estimated prevalence of IA in a general population of Korean adults. Our results revealed significant association between IA and other psychiatric disorders such as alcohol and nicotine use disorder, depressive disorder, and anxiety disorder.

Our findings of a 1.0% prevalence rate of IA among Korean internet users was similar to prevalence rate reported by a previous study conducted in Norway (1.0%), but higher than that estimated in a previous study conducted in the United States (0.3%). While more percentage of usual users used Internet retrieval to get useful information, addictive users were more likely to use social media and Internet gaming, consistent with the finding of Bakken et al.⁸ Addictive nature of Internet gaming has been previously recognized. In fact, Internet gaming disorder has been listed in DSM-5 as a condition for further research.² In addition, addictive qualities of social networking sites have been recently addressed.¹⁵¹⁶ They also require clinical attention.

Consistent with prior findings,⁸ addicted Internet users were more likely to be younger adults than usual users. This might be attributed to the fact that younger adults have easier access to technology. The finding of higher prevalence of IA in males is similar to that reported in prior studies.⁷¹⁷¹⁸ In addition, addicted Internet users were more likely to be unemployed, compatible with a positive association between IA and unsatisfactory financial status reported in a previous study.⁸ Excessive Internet use might provide a psychological escape from personal difficult situations such as unemployment. However, it might may further impede their social functioning as well.¹⁹

Our findings support the well-established associations between Internet addiction problem and a wide range of psychiatric disorders.³⁵⁷⁸ As found in prior studies among clinical sample and adolescents as well as young adult population, IA was associated with increased prevalence of substance use disorder,⁵⁸¹⁸ depressive disorder,^3781720 various types of anxiety disorder,³⁷ and obsessive-compulsive disorder.⁵⁷⁸

This cross-sectional study has several limitations. First, it cannot determine the causal relationship between IA and comorbid psychiatric disorders. In addition, IA was assessed via participants' self-reporting. Thus, further prospective studies using more structured clinical assessments for IA are needed to classify IA as a predictive factor or consequence of other psychiatric disorders. Despite such limitations, our results indicated that IA was closely linked with diverse psychiatric disorders in a general adult population in Korea.