The number of internet users is increasing rapidly in Korea. According to a report on the 2011 survey conducted by the Korea Internet Security Agency, the proportion of internet users among the Korean population over the age of 7 years was 44.7% in 2000 and the proportion among the population over the age of 3 years was 78.0% in 2011. In particular, 99.9% of teenagers used the internet in 2011.¹⁾

Adolescents get academic information easily and enjoy social interactions and leisure with low cost through the internet. Excessive use of the internet, however, can have deleterious effects on psychosocial development and behavior of adolescents.²⁾ Depression, anxiety, sleep disorders, and psychiatric disorders including attention deficit hyperactivity disorder are known to be related to internet addiction.^3-5)

The concept of 'internet addiction' as a discrete disease was first proposed in 1995 by Ivan Goldberg, a New York psychiatrist.⁶⁾ In 1996, Young⁷⁾ proposed criteria for 'internet addiction' patterned after the Diagnostic and Statistical Manual of Mental Disorders (DSM)-IV criteria for pathological gambling. Young's study dividing computer users into two groups, dependents and non-dependents, using the criteria, triggered controversy. Young said that on-line users could become addicted to the internet in the same way that others became addicted to drugs or alcohol, resulting in academic, social, and occupational impairment. In controversies, other investigators suggested several alternative terms including 'compulsive internet use,' 'pathological internet use,' 'problematic internet use,' and 'internet dependency,' and they proposed many diagnostic criteria.^8,9) In the draft diagnostic criteria for DSM-V released in 2010, the American Psychiatric Association considered internet addiction part of the 'behavioral addictions' category, but work group members decided there was insufficient research data to do so.¹⁰⁾

Despite the lack of definite diagnostic criteria, there is no doubt that internet addiction has become a serious public health problem in Asia, including Korea.^11-13) Therefore, studies concerning various factors related to internet addiction are conducted actively to understand and solve internet addiction phenomenon.^4,12) In light of behavioral addiction, researchers have been making efforts to find an association between internet addiction and other problem behaviors which can lead to addiction, such as alcohol drinking and drug abuse. Ko et al.¹⁴⁾ found that internet addiction was associated with problematic alcohol use, and psychosocial proneness towards problem behaviors is associated with internet addiction as well as problematic alcohol use in Taiwanese adolescents. Yen et al.¹⁵⁾ also reported that internet addiction was associated with harmful alcohol use in Taiwanese college students. June et al.¹⁶⁾ evaluated the relationships among internet addiction, cigarette smoking, and alcohol drinking in 1,592 Korean high school students. She concluded that smoking and drinking were interrelated with each other, but internet addiction was not related either with smoking or drinking.

The aim of our study is to identify the association between a high risk of internet addiction and problem behaviors which can lead to addiction, including cigarette smoking, alcohol drinking, drug abuse, and sexual intercourse among Korean adolescents.

A total of 73,238 students were enrolled in the current study, of which 3.0% were classified as having a 'high risk for internet addiction.' The socio-demographic characteristics and mental health status of the subjects were provided in Table 1. The percentage of boys was higher in the high risk group for internet addiction than in the low risk group. There were no differences in residing regions or type of schools between the two groups. The proportions of students who thought their academic performance would be in the lowest quintile and who thought their family economic status would be in the lowest quintile were higher in the high risk group for internet addiction. The percentages of parentless students and students who didn't live with family were also higher in the high risk group. The proportion of students whose parents completed college was lower in the high risk group than in the low risk group. Regarding mental health status, there were marked differences between the high risk group for internet addiction and the low risk group. The high risk group (57.1%) students reported that they experienced depression while the low risk group (36.8%) reported experiencing depression. The proportions of students who felt stressed and unhappy were 66.7% and 32.4% in the high risk group, and 43.1% and 12.9% in the low risk group, respectively.

Significantly higher proportions were shown in all of the problem behavior variables which included lifetime smoking experience, early smoking experience, current smoking, lifetime drinking experience, early drinking experience, current drinking, lifetime drug abuse experience, lifetime sexual intercourse experience, and early sexual intercourse experience among the high risk group for internet addiction regardless of gender (Tables 2, 3).

Table 4 showed the association between internet addiction and adolescent problem behaviors. The odds of lifetime smoking experience and early smoking experience were significantly higher among boys who were at high risk of internet addiction than for boys who were at low risk of internet addiction (adjusted odds ratio [AOR], 1.18; 95% confidence interval [CI], 1.01 to 1.38; AOR, 1.3; 95% CI, 1.04 to 1.62). Girls who were at high risk of internet addiction had significantly higher odds of lifetime smoking experience, early smoking experience, and current smoking than girls at low risk of internet addiction (AOR, 1.88; 95% CI, 1.48 to 2.39; AOR, 1.84; 95% CI, 1.33 to 2.55; AOR, 1.45; 95% CI, 1.03 to 2.04). The associations between the risk of internet addiction and lifetime drinking experience, early drinking experience and current drinking among boys were non-significant. However, among girls at high risk of internet addiction, the odds of lifetime drinking experience, early drinking experience, and current drinking were significantly higher than for girls at low risk of internet addiction (AOR, 1.33; 95% CI, 1.02 to 1.73; AOR, 1.31; 95% CI, 1.03 to 1.68; AOR, 1.43; 95% CI, 1.11 to 1.84). The odds of lifetime drug experience were significantly higher among boys and girls who were at high risk of internet addiction than for boys and girls who were at low risk of internet addiction (AOR, 4.3; 95% CI, 2.49 to 7.42; AOR, 4.21; 95% CI, 1.86 to 9.51). Boys who were at high risk of internet addiction had significantly higher odds of lifetime sexual intercourse experience and early sexual intercourse experience than boys who were at low risk of internet addiction (AOR, 1.36; 95% CI, 1.03 to 1.80; AOR, 2.48; 95% CI, 1.35 to 4.53). No significant association was observed between internet addiction and lifetime sexual intercourse experience and early sexual intercourse experience for girls.

Youths at high risk of internet addiction showed a high rate of cigarette smoking, alcohol drinking, drug abuse, and sexual intercourse experience in our study. These findings should not be interpreted as implying that internet addiction causes other problem behaviors among adolescents; however, it is likely that the same causal factors responsible for internet addiction increase the risk of internet addiction in adolescents engaging in other problem behaviors. It is consistent with Problem-Behavior Theory-psychosocial proneness of Jessor,²²⁾ which provides an explanation why adolescent problem behaviors and health risk behaviors cluster together. He suggested risk and protective factors for a "problem behavior syndrome" in 5 areas including biology/genetics, social environment, perceived environment, personality, and behavior, such as family history of alcoholism, poverty, conflicts with parents and friends, low self-esteem, and poor school work as risk factors, and high intelligence, quality school, models for conventional behavior, value on achievement and health, and church attendance as protective factors.^23,24)

The risk of internet addiction still had high associations with cigarette smoking, alcohol drinking, drug abuse, and sexual intercourse experiences even after adjusting for perceived academic performance, economic status, and family environment such as parents' level of education, stress, and depression. This shows that the association between internet addiction and smoking, drinking, drug abuse, and sexual intercourse experiences cannot be explained simply by environmental factors, social factors, and mental health status.

Ko et al.²⁵⁾ reported that Taiwanese adolescents with internet addiction were more likely to have substance use experience including tobacco, alcohol, or illicit drug use. He found that students addicted to the internet and students with substance use experience shared common personality characteristics more vulnerable to addiction. Similar findings among Greek adolescents were reported by Fisoun et al.²⁶⁾ These studies suggested that adolescents at high risk of internet addiction may have personalities vulnerable to any addiction and the personalities increase the risk for substance use and sexual intercourse which can lead to addiction. Although the exact neurobiological mechanism explaining the addictive personality has not been found, serotonin and dopamine are known to play an important role.²⁷⁾

Associations between internet addiction and problem behaviors differed between boys and girls in our study. Among girls, the risk of internet addiction was associated with all of the problem behavior variables except for sexual intercourse experience. However, in boys, sexual intercourse experience showed an association with the risk of internet addiction. This may be related to a report that boys with internet addiction visited websites with pornographic materials more frequently than girls with internet addiction.²⁸⁾ In boys, the risk of internet addiction was not associated with alcohol drinking. This could be related to the part of Korean culture that is more generous to men regarding alcohol consumption than to women. But more research is necessary to identify the cause of gender differences.

The limitations of this study include the cross-sectional study design, which cannot prove a causal or temporal relation between the risk of internet addiction and problem behaviors. Also, the responses were based on self-report, resulting in the possibility of response bias. As a clear diagnostic criteria for internet addiction has not been established yet, the validity of KS scale needs to be verified.

One of the key strengths of our study was its large nationwide representative sample of Korean middle and high school students. Most of the previous research on internet addiction performed in Korea had a problem with opportunistic sampling.^16,28,29) It was also meaningful that gender-stratified analysis was performed in this study.

In conclusion, it appears reasonable to consider concurrent problem behaviors, especially smoking, drinking, drug abuse, and sexual intercourse, when dealing with adolescents with a high risk of internet addiction.